Re: [PATCH] generic: prohibit fstrim on journalled filesystems with norecovery

[Filipe David Manana <fdmanana@apache.org>]

On Sat, Mar 23, 2019 at 12:35 AM Darrick J. Wong
<darrick.wong@oracle.com> wrote:
>
> From: Darrick J. Wong <darrick.wong@oracle.com>
>
> This test makes sure that we can't use stale unrecovered fs metadata to
> drive a DISCARD festival on a disk and thereby destroy user data by
> accident.

It would help to have listed the name of the patches that fix the
issues on xfs/ext4/btrfs, to make it faster.

>
> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>

Reviewed-by: Filipe Manana <fdmanana@suse.com>

Looks good, thanks for doing this. Just one question below.

> ---
>  tests/generic/714     |   61 +++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/generic/714.out |    4 +++
>  tests/generic/group   |    1 +
>  3 files changed, 66 insertions(+)
>  create mode 100755 tests/generic/714
>  create mode 100644 tests/generic/714.out
>
> diff --git a/tests/generic/714 b/tests/generic/714
> new file mode 100755
> index 00000000..1849a5e9
> --- /dev/null
> +++ b/tests/generic/714
> @@ -0,0 +1,61 @@
> +#! /bin/bash
> +# SPDX-License-Identifier: GPL-2.0+
> +# Copyright (c) 2019, Oracle and/or its affiliates.  All Rights Reserved.
> +#
> +# FS QA Test No. 714
> +#
> +# Ensure that we can't call fstrim on filesystems mounted norecovery, because
> +# FSTRIM implementations use free space metadata to drive the discard requests
> +# and we told the filesystem not to make sure the metadata are up to date.
> +
> +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 -rf $tmp.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# real QA test starts here
> +_supported_os Linux
> +_require_scratch
> +_require_fstrim
> +
> +rm -f $seqres.full
> +
> +_scratch_mkfs > $seqres.full 2>&1
> +_require_metadata_journaling $SCRATCH_DEV
> +
> +echo "fstrim on regular mount"
> +_scratch_mount >> $seqres.full 2>&1
> +$FSTRIM_PROG -v $SCRATCH_MNT >> $seqres.full 2>&1 || \
> +       _notrun "FSTRIM not supported"
> +_scratch_unmount
> +
> +echo "fstrim on ro mount"
> +_scratch_mount -o ro >> $seqres.full 2>&1
> +$FSTRIM_PROG -v $SCRATCH_MNT >> $seqres.full 2>&1
> +_scratch_unmount
> +
> +echo "fstrim on ro mount with no log replay"
> +norecovery="norecovery"
> +test $FSTYP = "btrfs" && norecovery=nologreplay
> +_scratch_mount -o ro,$norecovery >> $seqres.full 2>&1
> +$FSTRIM_PROG -v $SCRATCH_MNT >> $seqres.full 2>&1 && \
> +       echo "fstrim with unrecovered metadata just ate your filesystem"
> +_scratch_unmount
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/generic/714.out b/tests/generic/714.out
> new file mode 100644
> index 00000000..1158a2ff
> --- /dev/null
> +++ b/tests/generic/714.out
> @@ -0,0 +1,4 @@
> +QA output created by 714
> +fstrim on regular mount
> +fstrim on ro mount
> +fstrim on ro mount with no log replay
> diff --git a/tests/generic/group b/tests/generic/group
> index 2e4341fb..c2046293 100644
> --- a/tests/generic/group
> +++ b/tests/generic/group
> @@ -538,6 +538,7 @@
>  533 auto quick attr
>  534 auto quick log
>  535 auto quick log
> +714 auto trim

Any reason to not add the 'quick' group as well? It runs in 1 second
for me on btrfs, xfs and ext4 with a debug kernel.

>  940 auto quick clone punch
>  941 auto quick clone punch
>  942 auto quick clone punch



-- 
Filipe David Manana,

"Reasonable men adapt themselves to the world.
 Unreasonable men adapt the world to themselves.
 That's why all progress depends on unreasonable men."