Re: [PATCH 7/7] xfs/ext4: check negative inode size

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]



On Wed, Jan 04, 2017 at 05:05:20PM -0800, Darrick J. Wong wrote:
> Craft a malicious filesystem image with a negative inode size,
> then try to trigger a kernel DoS by appending data to the file.
> Ideally this should trigger verifier errors instead of hanging.
> 
> Signed-off-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
> ---
>  tests/shared/400     |   71 +++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/shared/400.out |    5 +++
>  tests/shared/401     |   71 +++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/shared/401.out |    5 +++
>  tests/shared/group   |    2 +
>  tests/xfs/400        |   72 ++++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/xfs/400.out    |    5 +++
>  tests/xfs/401        |   72 ++++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/xfs/401.out    |    5 +++
>  tests/xfs/group      |    2 +
>  10 files changed, 310 insertions(+)
>  create mode 100755 tests/shared/400
>  create mode 100644 tests/shared/400.out
>  create mode 100755 tests/shared/401
>  create mode 100644 tests/shared/401.out
>  create mode 100755 tests/xfs/400
>  create mode 100644 tests/xfs/400.out
>  create mode 100755 tests/xfs/401
>  create mode 100644 tests/xfs/401.out
> 
> 
> diff --git a/tests/shared/400 b/tests/shared/400
> new file mode 100755
> index 0000000..ba7bcda
> --- /dev/null
> +++ b/tests/shared/400
> @@ -0,0 +1,71 @@
> +#! /bin/bash
> +# FSQA Test No. 400
> +#
> +# Since loff_t is a signed type, it is invalid for a filesystem to load
> +# an inode with i_size = -1ULL.  Unfortunately, nobody checks this,
> +# which means that we can trivially DoS the VFS by creating such a file
> +# and appending to it.  This causes an integer overflow in the routines
> +# underlying writeback, which results in the kernel locking up.

How about adding some more descriptions here? Stating that this test is
testing buffered I/O, and adding some comments about testing direct I/O
to shared/401? The same to xfs/400 and xfs/401. Otherwise they look the
same just from the test description.

> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2017 Oracle, 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"
> +
> +PIDS=""
> +tmp=/tmp/$$
> +status=1	# failure is the default!
> +trap "_cleanup; exit \$status" 0 1 2 3 15
> +
> +_cleanup()
> +{
> +	rm -f $tmp.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# real QA test starts here
> +_supported_os Linux
> +_supported_fs ext2 ext3 ext4
> +_require_scratch_nocheck
> +_disable_dmesg_check

_require_command "$DEBUGFS_PROG" debugfs  and..

> +
> +rm -f $seqres.full
> +
> +echo "Format and mount"
> +_scratch_mkfs  >> $seqres.full 2>&1
> +_scratch_mount
> +
> +testdir=$SCRATCH_MNT
> +echo m > $testdir/a
> +
> +echo "Corrupt filesystem"
> +_scratch_unmount
> +debugfs -w -R "sif /a size -1" $SCRATCH_DEV >> $seqres.full 2>&1

$DEBUGFS_PROG here

> +
> +echo "Remount, try to append"
> +_scratch_mount
> +dd if=/dev/zero of=$testdir/a bs=512 count=1 oflag=append conv=notrunc >> $seqres.full 2>&1 || echo "Write did not succeed (ok)."
> +sync
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/shared/400.out b/tests/shared/400.out
> new file mode 100644
> index 0000000..ddf8a28
> --- /dev/null
> +++ b/tests/shared/400.out
> @@ -0,0 +1,5 @@
> +QA output created by 400
> +Format and mount
> +Corrupt filesystem
> +Remount, try to append
> +Write did not succeed (ok).
> diff --git a/tests/shared/401 b/tests/shared/401
> new file mode 100755
> index 0000000..d790381
> --- /dev/null
> +++ b/tests/shared/401
> @@ -0,0 +1,71 @@
> +#! /bin/bash
> +# FSQA Test No. 401
> +#
> +# Since loff_t is a signed type, it is invalid for a filesystem to load
> +# an inode with i_size = -1ULL.  Unfortunately, nobody checks this,
> +# which means that we can trivially DoS the VFS by creating such a file
> +# and appending to it.  This causes an integer overflow in the routines
> +# underlying writeback, which results in the kernel locking up.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2017 Oracle, 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"
> +
> +PIDS=""
> +tmp=/tmp/$$
> +status=1	# failure is the default!
> +trap "_cleanup; exit \$status" 0 1 2 3 15
> +
> +_cleanup()
> +{
> +	rm -f $tmp.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# real QA test starts here
> +_supported_os Linux
> +_supported_fs ext2 ext3 ext4
> +_require_scratch_nocheck
> +_disable_dmesg_check
> +
> +rm -f $seqres.full
> +
> +echo "Format and mount"
> +_scratch_mkfs  >> $seqres.full 2>&1
> +_scratch_mount
> +
> +testdir=$SCRATCH_MNT
> +echo m > $testdir/a
> +
> +echo "Corrupt filesystem"
> +_scratch_unmount
> +debugfs -w -R "sif /a size 0xFFFFFFFFFFFFFE00" $SCRATCH_DEV >> $seqres.full 2>&1

Better to have your explanation to 0XFFFFFFFFFFFFFE00 as comments. (Same
to xfs/400 and xfs/401.)

"The 0xFFFFFFFFFFFFFE00 rounds the file size down to a multiple of 512
so that we can do the directio"

Thanks,
Eryu

> +
> +echo "Remount, try to append"
> +_scratch_mount
> +dd if=/dev/zero of=$testdir/a bs=512 count=1 oflag=direct,append conv=notrunc >> $seqres.full 2>&1 || echo "Write did not succeed (ok)."
> +sync
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/shared/401.out b/tests/shared/401.out
> new file mode 100644
> index 0000000..0d45add
> --- /dev/null
> +++ b/tests/shared/401.out
> @@ -0,0 +1,5 @@
> +QA output created by 401
> +Format and mount
> +Corrupt filesystem
> +Remount, try to append
> +Write did not succeed (ok).
> diff --git a/tests/shared/group b/tests/shared/group
> index 55bb594..64d2d2f 100644
> --- a/tests/shared/group
> +++ b/tests/shared/group
> @@ -13,3 +13,5 @@
>  272 auto enospc rw
>  289 auto quick
>  298 auto trim
> +400 dangerous_fuzzers
> +401 dangerous_fuzzers
> diff --git a/tests/xfs/400 b/tests/xfs/400
> new file mode 100755
> index 0000000..eed8fdf
> --- /dev/null
> +++ b/tests/xfs/400
> @@ -0,0 +1,72 @@
> +#! /bin/bash
> +# FSQA Test No. 400
> +#
> +# Since loff_t is a signed type, it is invalid for a filesystem to load
> +# an inode with i_size = -1ULL.  Unfortunately, nobody checks this,
> +# which means that we can trivially DoS the VFS by creating such a file
> +# and appending to it.  This causes an integer overflow in the routines
> +# underlying writeback, which results in the kernel locking up.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2017 Oracle, 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"
> +
> +PIDS=""
> +tmp=/tmp/$$
> +status=1	# failure is the default!
> +trap "_cleanup; exit \$status" 0 1 2 3 15
> +
> +_cleanup()
> +{
> +	rm -f $tmp.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# real QA test starts here
> +_supported_os Linux
> +_supported_fs xfs
> +_require_scratch_nocheck
> +_disable_dmesg_check
> +
> +rm -f $seqres.full
> +
> +echo "Format and mount"
> +_scratch_mkfs  >> $seqres.full 2>&1
> +_scratch_mount
> +
> +testdir=$SCRATCH_MNT
> +echo m > $testdir/a
> +inum=$(stat -c "%i" $testdir/a)
> +
> +echo "Corrupt filesystem"
> +_scratch_unmount
> +_scratch_xfs_db -x -c "inode ${inum}" -c 'write core.size -- -1' >> $seqres.full
> +
> +echo "Remount, try to append"
> +_scratch_mount
> +dd if=/dev/zero of=$testdir/a bs=512 count=1 oflag=append conv=notrunc >> $seqres.full 2>&1 || echo "Write did not succeed (ok)."
> +sync
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/xfs/400.out b/tests/xfs/400.out
> new file mode 100644
> index 0000000..ddf8a28
> --- /dev/null
> +++ b/tests/xfs/400.out
> @@ -0,0 +1,5 @@
> +QA output created by 400
> +Format and mount
> +Corrupt filesystem
> +Remount, try to append
> +Write did not succeed (ok).
> diff --git a/tests/xfs/401 b/tests/xfs/401
> new file mode 100755
> index 0000000..2be684a
> --- /dev/null
> +++ b/tests/xfs/401
> @@ -0,0 +1,72 @@
> +#! /bin/bash
> +# FSQA Test No. 401
> +#
> +# Since loff_t is a signed type, it is invalid for a filesystem to load
> +# an inode with i_size = -1ULL.  Unfortunately, nobody checks this,
> +# which means that we can trivially DoS the VFS by creating such a file
> +# and appending to it.  This causes an integer overflow in the routines
> +# underlying writeback, which results in the kernel locking up.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2017 Oracle, 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"
> +
> +PIDS=""
> +tmp=/tmp/$$
> +status=1	# failure is the default!
> +trap "_cleanup; exit \$status" 0 1 2 3 15
> +
> +_cleanup()
> +{
> +	rm -f $tmp.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# real QA test starts here
> +_supported_os Linux
> +_supported_fs xfs
> +_require_scratch_nocheck
> +_disable_dmesg_check
> +
> +rm -f $seqres.full
> +
> +echo "Format and mount"
> +_scratch_mkfs  >> $seqres.full 2>&1
> +_scratch_mount
> +
> +testdir=$SCRATCH_MNT
> +echo m > $testdir/a
> +inum=$(stat -c "%i" $testdir/a)
> +
> +echo "Corrupt filesystem"
> +_scratch_unmount
> +_scratch_xfs_db -x -c "inode ${inum}" -c 'write core.size -- -512' >> $seqres.full
> +
> +echo "Remount, try to append"
> +_scratch_mount
> +dd if=/dev/zero of=$testdir/a bs=512 count=1 oflag=direct,append conv=notrunc >> $seqres.full 2>&1 || echo "Write did not succeed (ok)."
> +sync
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/xfs/401.out b/tests/xfs/401.out
> new file mode 100644
> index 0000000..0d45add
> --- /dev/null
> +++ b/tests/xfs/401.out
> @@ -0,0 +1,5 @@
> +QA output created by 401
> +Format and mount
> +Corrupt filesystem
> +Remount, try to append
> +Write did not succeed (ok).
> diff --git a/tests/xfs/group b/tests/xfs/group
> index c237b50..10ba27b 100644
> --- a/tests/xfs/group
> +++ b/tests/xfs/group
> @@ -334,3 +334,5 @@
>  345 auto quick clone
>  346 auto quick clone
>  347 auto quick clone
> +400 dangerous_fuzzers
> +401 dangerous_fuzzers
> 
--
To unsubscribe from this list: send the line "unsubscribe fstests" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [Linux Filesystems Development]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux