linux-ext4 - Re: [PATCH 09/12] xfs/521: require a real SCRATCH

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20251210195427.GE94594@frogsfrogsfrogs>
Date: Wed, 10 Dec 2025 11:54:27 -0800
From: "Darrick J. Wong" <djwong@...nel.org>
To: Christoph Hellwig <hch@....de>
Cc: Zorro Lang <zlang@...nel.org>, Anand Jain <anand.jain@...cle.com>,
	Filipe Manana <fdmanana@...e.com>, fstests@...r.kernel.org,
	linux-ext4@...r.kernel.org, linux-xfs@...r.kernel.org
Subject: Re: [PATCH 09/12] xfs/521: require a real SCRATCH_RTDEV

On Wed, Dec 10, 2025 at 06:46:55AM +0100, Christoph Hellwig wrote:
> Require a real SCRATCH_RTDEV instead of faking one up using a loop
> device, as otherwise the options specified in MKFS_OPTIONS might
> not actually work the configuration.
> 
> Signed-off-by: Christoph Hellwig <hch@....de>
> ---
>  tests/xfs/521     | 33 ++++++++-------------------------
>  tests/xfs/521.out |  1 -
>  2 files changed, 8 insertions(+), 26 deletions(-)
> 
> diff --git a/tests/xfs/521 b/tests/xfs/521
> index 0da05a55a276..5cd6649c50c7 100755
> --- a/tests/xfs/521
> +++ b/tests/xfs/521
> @@ -16,34 +16,16 @@
>  . ./common/preamble
>  _begin_fstest auto quick realtime growfs
>  
> -# Override the default cleanup function.
> -_cleanup()
> -{
> -	cd /
> -	_scratch_unmount >> $seqres.full 2>&1
> -	[ -n "$rt_loop_dev" ] && _destroy_loop_device $rt_loop_dev
> -	rm -f $tmp.* $TEST_DIR/$seq.rtvol
> -}
> -
> -# Import common functions.
>  . ./common/filter
>  
> -# Note that we don't _require_realtime because we synthesize a rt volume
> -# below.
> -_require_scratch_nocheck
> -_require_no_large_scratch_dev
> -
> -echo "Create fake rt volume"
> -truncate -s 400m $TEST_DIR/$seq.rtvol
> -rt_loop_dev=$(_create_loop_device $TEST_DIR/$seq.rtvol)
> +_require_realtime
> +_require_scratch
>  
>  echo "Format and mount 100m rt volume"
> -export USE_EXTERNAL=yes
> -export SCRATCH_RTDEV=$rtdev
>  _scratch_mkfs -r size=100m > $seqres.full
> -_try_scratch_mount || _notrun "Could not mount scratch with synthetic rt volume"
> +_scratch_mount
>  
> -# zoned file systems only support zoned size-rounded RT device sizes
> +# zoned file systems only support zone-size aligned RT device sizes
>  _require_xfs_scratch_non_zoned
>  
>  testdir=$SCRATCH_MNT/test-$seq
> @@ -58,7 +40,10 @@ echo "Create some files"
>  _pwrite_byte 0x61 0 1m $testdir/original >> $seqres.full
>  
>  echo "Grow fs"
> -$XFS_GROWFS_PROG $SCRATCH_MNT 2>&1 |  _filter_growfs >> $seqres.full
> +# growfs expects sizes in FSB units
> +fsbsize=$(_get_block_size $SCRATCH_MNT)
> +$XFS_GROWFS_PROG -R $((400 * 1024 * 1024 / fsbsize)) $SCRATCH_MNT 2>&1 | \

Hrmm, I wonder if this test should check that SCRATCH_RTDEV is at least
400m?  But I guess the old code didn't check that the loop file doesn't
ENOSPC (which is another good reason to get rid of the fakery) so...

Reviewed-by: "Darrick J. Wong" <djwong@...nel.org>

--D

> +	  _filter_growfs >> $seqres.full
>  _scratch_cycle_mount
>  
>  echo "Recheck 400m rt volume stats"
> @@ -73,8 +58,6 @@ echo "Check filesystem"
>  _check_scratch_fs
>  
>  _scratch_unmount
> -_destroy_loop_device $rt_loop_dev
> -unset rt_loop_dev
>  
>  # success, all done
>  status=0
> diff --git a/tests/xfs/521.out b/tests/xfs/521.out
> index 007ab92c6db2..afd18bb0dc99 100644
> --- a/tests/xfs/521.out
> +++ b/tests/xfs/521.out
> @@ -1,5 +1,4 @@
>  QA output created by 521
> -Create fake rt volume
>  Format and mount 100m rt volume
>  Check rt volume stats
>  Create some files
> -- 
> 2.47.3
> 
>