public inbox for fstests@vger.kernel.org
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <djwong@kernel.org>
To: Amir Goldstein <amir73il@gmail.com>
Cc: zlang@redhat.com, neal@gompa.dev, fstests@vger.kernel.org,
	linux-ext4@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	joannelkoong@gmail.com, bernd@bsbernd.com
Subject: Re: [PATCH 01/33] misc: adapt tests to handle the fuse ext[234] drivers
Date: Wed, 5 Nov 2025 14:53:55 -0800	[thread overview]
Message-ID: <20251105225355.GC196358@frogsfrogsfrogs> (raw)
In-Reply-To: <CAOQ4uxj7yaX5qLEs4BOJBJwybkHzv8WmNsUt0w_zehueOLLP9A@mail.gmail.com>

On Thu, Oct 30, 2025 at 10:51:06AM +0100, Amir Goldstein wrote:
> On Wed, Oct 29, 2025 at 2:22 AM Darrick J. Wong <djwong@kernel.org> wrote:
> >
> > From: Darrick J. Wong <djwong@kernel.org>
> >
> > It would be useful to be able to run fstests against the userspace
> > ext[234] driver program fuse2fs.  A convention (at least on Debian)
> > seems to be to install fuse drivers as /sbin/mount.fuse.XXX so that
> > users can run "mount -t fuse.XXX" to start a fuse driver for a
> > disk-based filesystem type XXX.
> >
> > Therefore, we'll adopt the practice of setting FSTYP=fuse.ext4 to
> > test ext4 with fuse2fs.  Change all the library code as needed to handle
> > this new type alongside all the existing ext[234] checks, which seems a
> > little cleaner than FSTYP=fuse FUSE_SUBTYPE=ext4, which also would
> > require even more treewide cleanups to work properly because most
> > fstests code switches on $FSTYP alone.
> >
> 
> I agree that FSTYP=fuse.ext4 is cleaner than
> FSTYP=fuse FUSE_SUBTYPE=ext4
> but it is not extendable to future (e.g. fuse.xfs)
> and it is still a bit ugly.
> 
> Consider:
> FSTYP=fuse.ext4
> MKFSTYP=ext4
> 
> I think this is the correct abstraction -
> fuse2fs/ext4 are formatted that same and mounted differently
> 
> See how some of your patch looks nicer and naturally extends to
> the imaginary fuse.xfs...

Maybe I'd rather do it the other way around for fuse4fs:

FSTYP=ext4
MOUNT_FSTYP=fuse.ext4

(obviously, MOUNT_FSTYP=$FSTYP if the test runner hasn't overridden it)

Where $MOUNT_FSTYP is what you pass to mount -t and what you'd see in
/proc/mounts.  The only weirdness with that is that some of the helpers
will end up with code like:

	case $FSTYP in
	ext4)
		# do ext4 stuff
		;;
	esac

	case $MOUNT_FSTYP in
	fuse.ext4)
		# do fuse4fs stuff that overrides ext4
		;;
	esac

which would be a little weird.

_scratch_mount would end up with:

	$MOUNT_PROG -t $MOUNT_FSTYP ...

and detecting it would be

	grep -q -w $MOUNT_FSTYP /proc/mounts || _fail "booooo"

Hrm?

--D

> 
> > Signed-off-by: "Darrick J. Wong" <djwong@kernel.org>
> > ---
> >  check             |   24 +++++++++++++++++-------
> >  common/casefold   |    4 ++++
> >  common/config     |   11 ++++++++---
> >  common/defrag     |    2 +-
> >  common/encrypt    |   16 ++++++++--------
> >  common/log        |   10 +++++-----
> >  common/populate   |   14 +++++++-------
> >  common/quota      |    9 +++++++++
> >  common/rc         |   50 +++++++++++++++++++++++++++++---------------------
> >  common/report     |    2 +-
> >  common/verity     |    8 ++++----
> >  tests/generic/020 |    2 +-
> >  tests/generic/067 |    2 +-
> >  tests/generic/441 |    2 +-
> >  tests/generic/496 |    2 +-
> >  tests/generic/621 |    2 +-
> >  tests/generic/740 |    2 +-
> >  tests/generic/746 |    4 ++--
> >  tests/generic/765 |    4 ++--
> >  19 files changed, 103 insertions(+), 67 deletions(-)
> >
> >
> > diff --git a/check b/check
> > index 9bb80a22440f97..81cd03f73ce155 100755
> > --- a/check
> > +++ b/check
> > @@ -140,12 +140,25 @@ get_sub_group_list()
> >         echo $grpl
> >  }
> >
> > +get_group_dirs()
> > +{
> > +       local fsgroup="$FSTYP"
> > +
> > +       case "$FSTYP" in
> > +       ext2|ext3|fuse.ext[234])
> > +               fsgroup=ext4
> > +               ;;
> > +       esac
> > +
> > +       echo $SRC_GROUPS
> > +       echo $fsgroup
> > +}
> > +
> >  get_group_list()
> >  {
> >         local grp=$1
> >         local grpl=""
> >         local sub=$(dirname $grp)
> > -       local fsgroup="$FSTYP"
> >
> >         if [ -n "$sub" -a "$sub" != "." -a -d "$SRC_DIR/$sub" ]; then
> >                 # group is given as <subdir>/<group> (e.g. xfs/quick)
> > @@ -154,10 +167,7 @@ get_group_list()
> >                 return
> >         fi
> >
> > -       if [ "$FSTYP" = ext2 -o "$FSTYP" = ext3 ]; then
> > -           fsgroup=ext4
> > -       fi
> > -       for d in $SRC_GROUPS $fsgroup; do
> > +       for d in $(get_group_dirs); do
> >                 if ! test -d "$SRC_DIR/$d" ; then
> >                         continue
> >                 fi
> > @@ -171,7 +181,7 @@ get_group_list()
> >  get_all_tests()
> >  {
> >         touch $tmp.list
> > -       for d in $SRC_GROUPS $FSTYP; do
> > +       for d in $(get_group_dirs); do
> >                 if ! test -d "$SRC_DIR/$d" ; then
> >                         continue
> >                 fi
> > @@ -387,7 +397,7 @@ if [ -n "$FUZZ_REWRITE_DURATION" ]; then
> >  fi
> >
> >  if [ -n "$subdir_xfile" ]; then
> > -       for d in $SRC_GROUPS $FSTYP; do
> > +       for d in $(get_group_dirs); do
> >                 [ -f $SRC_DIR/$d/$subdir_xfile ] || continue
> >                 for f in `sed "s/#.*$//" $SRC_DIR/$d/$subdir_xfile`; do
> >                         exclude_tests+=($d/$f)
> > diff --git a/common/casefold b/common/casefold
> > index 2aae5e5e6c8925..fcdb4d210028ac 100644
> > --- a/common/casefold
> > +++ b/common/casefold
> > @@ -6,6 +6,10 @@
> >  _has_casefold_kernel_support()
> >  {
> >         case $FSTYP in
> > +       fuse.ext[234])
> > +               # fuse2fs does not support casefolding
> > +               false
> > +               ;;
> 
> This would not be needed
> 
> >         ext4)
> >                 test -f '/sys/fs/ext4/features/casefold'
> >                 ;;
> > diff --git a/common/config b/common/config
> > index 7fa97319d7d0ca..0cd2b33c4ade40 100644
> > --- a/common/config
> > +++ b/common/config
> > @@ -386,6 +386,11 @@ _common_mount_opts()
> >         overlay)
> >                 echo $OVERLAY_MOUNT_OPTIONS
> >                 ;;
> > +       fuse.ext[234])
> > +               # fuse sets up secure defaults, so we must explicitly tell
> > +               # fuse2fs to use the more relaxed kernel access behaviors.
> > +               echo "-o kernel $EXT_MOUNT_OPTIONS"
> > +               ;;
> >         ext2|ext3|ext4)
> >                 # acls & xattrs aren't turned on by default on ext$FOO
> >                 echo "-o acl,user_xattr $EXT_MOUNT_OPTIONS"
> > @@ -472,7 +477,7 @@ _mkfs_opts()
> >  _fsck_opts()
> >  {
> >         case $FSTYP in
> 
> This would obviously be $MKFSTYP with no further changes
> 
> > -       ext2|ext3|ext4)
> > +       ext2|ext3|fuse.ext[234]|ext4)
> >                 export FSCK_OPTIONS="-nf"
> >                 ;;
> >         reiser*)
> > @@ -514,11 +519,11 @@ _source_specific_fs()
> >
> >                 . ./common/btrfs
> >                 ;;
> > -       ext4)
> > +       fuse.ext4|ext4)
> >                 [ "$MKFS_EXT4_PROG" = "" ] && _fatal "mkfs.ext4 not found"
> >                 . ./common/ext4
> >                 ;;
> > -       ext2|ext3)
> > +       ext2|ext3|fuse.ext[23])
> >                 . ./common/ext4
> 
> same here
> 
> >                 ;;
> >         f2fs)
> > diff --git a/common/defrag b/common/defrag
> > index 055d0d0e9182c5..c054e62bde6f4d 100644
> > --- a/common/defrag
> > +++ b/common/defrag
> > @@ -12,7 +12,7 @@ _require_defrag()
> >          _require_xfs_io_command "falloc"
> >          DEFRAG_PROG="$XFS_FSR_PROG"
> >         ;;
> > -    ext4)
> > +    fuse.ext4|ext4)
> >         testfile="$TEST_DIR/$$-test.defrag"
> >         donorfile="$TEST_DIR/$$-donor.defrag"
> >         bsize=`_get_block_size $TEST_DIR`
> 
> and here
> 
> > diff --git a/common/encrypt b/common/encrypt
> > index f2687631b214cf..4fa7b6853fd461 100644
> > --- a/common/encrypt
> > +++ b/common/encrypt
> > @@ -191,7 +191,7 @@ _require_hw_wrapped_key_support()
> >  _scratch_mkfs_encrypted()
> >  {
> >         case $FSTYP in
> > -       ext4|f2fs)
> > +       fuse.ext4|ext4|f2fs)
> >                 _scratch_mkfs -O encrypt
> >                 ;;
> 
> and here
> 
> >         ubifs)
> > @@ -210,7 +210,7 @@ _scratch_mkfs_encrypted()
> >  _scratch_mkfs_sized_encrypted()
> >  {
> >         case $FSTYP in
> > -       ext4|f2fs)
> > +       fuse.ext4|ext4|f2fs)
> >                 MKFS_OPTIONS="$MKFS_OPTIONS -O encrypt" _scratch_mkfs_sized $*
> >                 ;;
> 
> and here... I think you got my point.
> 
> Thanks,
> Amir.
> 

  reply	other threads:[~2025-11-05 22:53 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20251029002755.GK6174@frogsfrogsfrogs>
2025-10-29  0:42 ` [PATCHSET v6] fstests: support ext4 fuse testing Darrick J. Wong
2025-10-29  1:20   ` [PATCH 01/33] misc: adapt tests to handle the fuse ext[234] drivers Darrick J. Wong
2025-10-30  9:51     ` Amir Goldstein
2025-11-05 22:53       ` Darrick J. Wong [this message]
2025-11-06  8:58         ` Amir Goldstein
2025-11-06 23:12           ` Darrick J. Wong
2025-11-07  7:50             ` Amir Goldstein
2025-11-07  7:08         ` Zorro Lang
2025-10-29  1:20   ` [PATCH 02/33] generic/740: don't run this test for fuse ext* implementations Darrick J. Wong
2025-10-30  9:59     ` Amir Goldstein
2025-11-05 22:56       ` Darrick J. Wong
2025-11-06  9:02         ` Amir Goldstein
2025-10-29  1:21   ` [PATCH 03/33] ext/052: use popdir.pl for much faster directory creation Darrick J. Wong
2025-10-29  1:21   ` [PATCH 04/33] common/rc: skip test if swapon doesn't work Darrick J. Wong
2025-11-12  6:35     ` Baokun Li
2025-11-12 18:26       ` Darrick J. Wong
2025-11-12 20:05         ` Theodore Ts'o
2025-11-12 22:29           ` [PATCH v6.1 " Darrick J. Wong
2025-11-13  1:51             ` Baokun Li
2025-11-13 15:52             ` Theodore Ts'o
2025-10-29  1:21   ` [PATCH 05/33] common/rc: streamline _scratch_remount Darrick J. Wong
2025-10-29  1:21   ` [PATCH 06/33] ext/039: require metadata journalling Darrick J. Wong
2025-10-29  1:22   ` [PATCH 07/33] populate: don't check for htree directories on fuse.ext4 Darrick J. Wong
2025-10-29  1:22   ` [PATCH 08/33] misc: convert _scratch_mount -o remount to _scratch_remount Darrick J. Wong
2025-10-29  1:22   ` [PATCH 09/33] misc: use explicitly $FSTYP'd mount calls Darrick J. Wong
2025-10-29  1:23   ` [PATCH 10/33] common/ext4: explicitly format with $FSTYP Darrick J. Wong
2025-10-29  1:23   ` [PATCH 11/33] tests/ext*: refactor open-coded _scratch_mkfs_sized calls Darrick J. Wong
2025-10-29  1:23   ` [PATCH 12/33] generic/732: disable for fuse.ext4 Darrick J. Wong
2025-10-29  1:23   ` [PATCH 13/33] defrag: fix ext4 defrag ioctl test Darrick J. Wong
2025-10-29  1:24   ` [PATCH 14/33] misc: explicitly require online resize support Darrick J. Wong
2025-10-29  1:24   ` [PATCH 15/33] ext4/004: disable for fuse2fs Darrick J. Wong
2025-10-29  1:24   ` [PATCH 16/33] generic/679: " Darrick J. Wong
2025-10-29  1:24   ` [PATCH 17/33] ext4/045: don't run the long dirent test on fuse2fs Darrick J. Wong
2025-10-29  1:25   ` [PATCH 18/33] generic/338: skip test if we can't mount with strictatime Darrick J. Wong
2025-10-29  1:25   ` [PATCH 19/33] generic/563: fuse doesn't support cgroup-aware writeback accounting Darrick J. Wong
2025-10-29  1:25   ` [PATCH 20/33] misc: use a larger buffer size for pwrites Darrick J. Wong
2025-10-29  1:25   ` [PATCH 21/33] ext4/046: don't run this test if dioread_nolock not supported Darrick J. Wong
2025-10-29  1:26   ` [PATCH 22/33] generic/631: don't run test if we can't mount overlayfs Darrick J. Wong
2025-10-30 11:35     ` Amir Goldstein
2025-11-05 23:12       ` Darrick J. Wong
2025-11-06  9:23         ` Amir Goldstein
2025-11-06 16:02           ` Darrick J. Wong
2025-10-29  1:26   ` [PATCH 23/33] generic/{409,410,411,589}: check for stacking mount support Darrick J. Wong
2025-10-30 10:25     ` Amir Goldstein
2025-11-05 22:58       ` Darrick J. Wong
2025-10-29  1:26   ` [PATCH 24/33] generic: add _require_hardlinks to tests that require hardlinks Darrick J. Wong
2025-10-29  1:26   ` [PATCH 25/33] ext4/001: check for fiemap support Darrick J. Wong
2025-10-29  1:27   ` [PATCH 26/33] generic/622: check that strictatime/lazytime actually work Darrick J. Wong
2025-10-29  1:27   ` [PATCH 27/33] generic/050: skip test because fuse2fs doesn't have stable output Darrick J. Wong
2025-10-30 10:05     ` Amir Goldstein
2025-11-05 23:02       ` Darrick J. Wong
2025-10-29  1:27   ` [PATCH 28/33] generic/405: don't stall on mkfs asking for input Darrick J. Wong
2025-10-29  1:27   ` [PATCH 29/33] ext4/006: fix this test Darrick J. Wong
2025-10-29  1:28   ` [PATCH 30/33] ext4/009: fix ENOSPC errors Darrick J. Wong
2025-10-29  1:28   ` [PATCH 31/33] ext4/022: enabl Darrick J. Wong
2025-10-29  6:03     ` Darrick J. Wong
2025-10-29  1:28   ` [PATCH 32/33] generic/730: adapt test for fuse filesystems Darrick J. Wong
2025-10-29  1:29   ` [PATCH 33/33] fuse2fs: hack around weird corruption problems Darrick J. Wong
2025-10-29  9:35   ` [PATCHSET v6] fstests: support ext4 fuse testing Christoph Hellwig
2025-10-29 23:52     ` Darrick J. Wong

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20251105225355.GC196358@frogsfrogsfrogs \
    --to=djwong@kernel.org \
    --cc=amir73il@gmail.com \
    --cc=bernd@bsbernd.com \
    --cc=fstests@vger.kernel.org \
    --cc=joannelkoong@gmail.com \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=neal@gompa.dev \
    --cc=zlang@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox