From: "Darrick J. Wong" <djwong@kernel.org>
To: Zorro Lang <zlang@redhat.com>
Cc: fstests@vger.kernel.org, linux-ext4@vger.kernel.org
Subject: Re: [PATCH 1/1] common: test statfs reporting with project quota
Date: Tue, 4 Mar 2025 10:26:59 -0800 [thread overview]
Message-ID: <20250304182659.GC2803740@frogsfrogsfrogs> (raw)
In-Reply-To: <20250302162610.y4l453sjzlw75agr@dell-per750-06-vm-08.rhts.eng.pek2.redhat.com>
On Mon, Mar 03, 2025 at 12:26:10AM +0800, Zorro Lang wrote:
> On Tue, Feb 18, 2025 at 05:03:39PM -0800, Darrick J. Wong wrote:
> > From: Darrick J. Wong <djwong@kernel.org>
> >
> > Create a test to check that statfs on a directory tree with a project
> > quota will report the quota limit and available blocks; and that the
> > available blocks reported doesn't exceed that of the whole filesystem.
> >
> > Signed-off-by: "Darrick J. Wong" <djwong@kernel.org>
> > ---
> > tests/generic/1955 | 114 ++++++++++++++++++++++++++++++++++++++++++++++++
> > tests/generic/1955.out | 13 +++++
> > 2 files changed, 127 insertions(+)
> > create mode 100755 tests/generic/1955
> > create mode 100644 tests/generic/1955.out
> >
> >
> > diff --git a/tests/generic/1955 b/tests/generic/1955
> > new file mode 100755
> > index 00000000000000..e431b3c4e3fd5d
> > --- /dev/null
> > +++ b/tests/generic/1955
> > @@ -0,0 +1,114 @@
> > +#! /bin/bash
> > +# SPDX-License-Identifier: GPL-2.0
> > +# Copyright (c) 2024-2025 Oracle. All Rights Reserved.
> > +#
> > +# FS QA Test No. 1955
> > +#
> > +# Make sure that statfs reporting works when project quotas are set on a
> > +# directory tree.
> > +#
> > +. ./common/preamble
> > +_begin_fstest auto quota
> > +
> > +_fixed_by_git_commit kernel XXXXXXXXXXXXXX \
> > + "xfs: don't over-report free space or inodes in statvfs"
>
>
> Ext4 fails on this test [1], is it a known issue of ext4 besides xfs?
Yeah, known issue on ext4 (and probably f2fs too).
--D
> Thanks,
> Zorro
>
> [1]
> FSTYP -- ext4
> PLATFORM -- Linux/aarch64 hpe-apollo-cn99xx-14-vm-28 6.14.0-rc4+ #1 SMP PREEMPT_DYNAMIC Sat Mar 1 16:57:43 EST 2025
> MKFS_OPTIONS -- -F /dev/vda3
> MOUNT_OPTIONS -- -o acl,user_xattr -o context=system_u:object_r:root_t:s0 /dev/vda3 /mnt/xfstests/scratch
>
> generic/762 - output mismatch (see /var/lib/xfstests/results//generic/762.out.bad)
> --- tests/generic/762.out 2025-03-01 17:23:52.961878242 -0500
> +++ /var/lib/xfstests/results//generic/762.out.bad 2025-03-01 19:13:03.026588012 -0500
> @@ -6,8 +6,10 @@
> root blocks2 is in range
> dir blocks2 is in range
> root bavail2 is in range
> -dir bavail2 is in range
> +dir bavail2 has value of 1821553
> +dir bavail2 is NOT in range 360666.9 .. 367953.1
> root blocks3 is in range
> ...
> (Run 'diff -u /var/lib/xfstests/tests/generic/762.out /var/lib/xfstests/results//generic/762.out.bad' to see the entire diff)
>
> HINT: You _MAY_ be missing kernel fix:
> XXXXXXXXXXXXXX xfs: don't over-report free space or inodes in statvfs
>
> Ran: generic/762
> Failures: generic/762
> Failed 1 of 1 tests
>
> > +
> > +. ./common/filter
> > +. ./common/quota
> > +
> > +_require_quota
> > +_require_scratch
> > +_require_xfs_io_command 'chproj'
> > +_require_xfs_io_command "falloc"
> > +
> > +_scratch_mkfs >$seqres.full 2>&1
> > +_scratch_enable_pquota
> > +_qmount_option "prjquota"
> > +_qmount
> > +_force_vfs_quota_testing $SCRATCH_MNT
> > +_require_prjquota $SCRATCH_DEV
> > +
> > +mkdir $SCRATCH_MNT/dir
> > +
> > +bsize() {
> > + $XFS_IO_PROG -c 'statfs' $1 | grep f_bsize | awk '{print $3}'
> > +}
> > +
> > +blocks() {
> > + $XFS_IO_PROG -c 'statfs' $1 | grep f_blocks | awk '{print $3}'
> > +}
> > +
> > +bavail() {
> > + $XFS_IO_PROG -c 'statfs' $1 | grep f_bavail | awk '{print $3}'
> > +}
> > +
> > +bsize=$(bsize $SCRATCH_MNT)
> > +orig_bavail=$(bavail $SCRATCH_MNT)
> > +orig_blocks=$(blocks $SCRATCH_MNT)
> > +
> > +# Set a project quota limit of half the free space, make sure both report the
> > +# same number of blocks
> > +pquot_limit=$(( orig_bavail / 2 ))
> > +setquota -P 55 0 $((pquot_limit * bsize / 1024))K 0 0 $SCRATCH_DEV
> > +$XFS_IO_PROG -c 'chproj 55' -c 'chattr +P' $SCRATCH_MNT/dir
> > +
> > +# check statfs reporting
> > +fs_blocks=$(blocks $SCRATCH_MNT)
> > +dir_blocks=$(blocks $SCRATCH_MNT/dir)
> > +
> > +_within_tolerance "root blocks1" $fs_blocks $orig_blocks 1% -v
> > +_within_tolerance "dir blocks1" $dir_blocks $pquot_limit 1% -v
> > +
> > +fs_bavail=$(bavail $SCRATCH_MNT)
> > +expected_dir_bavail=$pquot_limit
> > +dir_bavail=$(bavail $SCRATCH_MNT/dir)
> > +
> > +_within_tolerance "root bavail1" $fs_bavail $orig_bavail 1% -v
> > +_within_tolerance "dir bavail1" $dir_bavail $expected_dir_bavail 1% -v
> > +
> > +# use up most of the free space in the filesystem
> > +rem_free=$(( orig_bavail / 10 )) # bsize blocks
> > +fallocate -l $(( (orig_bavail - rem_free) * bsize )) $SCRATCH_MNT/a
> > +
> > +if [ $rem_free -gt $pquot_limit ]; then
> > + echo "rem_free $rem_free greater than pquot_limit $pquot_limit??"
> > +fi
> > +
> > +# check statfs reporting
> > +fs_blocks=$(blocks $SCRATCH_MNT)
> > +dir_blocks=$(blocks $SCRATCH_MNT/dir)
> > +
> > +_within_tolerance "root blocks2" $fs_blocks $orig_blocks 1% -v
> > +_within_tolerance "dir blocks2" $dir_blocks $pquot_limit 1% -v
> > +
> > +fs_bavail=$(bavail $SCRATCH_MNT)
> > +dir_bavail=$(bavail $SCRATCH_MNT/dir)
> > +
> > +_within_tolerance "root bavail2" $fs_bavail $rem_free 1% -v
> > +_within_tolerance "dir bavail2" $dir_bavail $rem_free 1% -v
> > +
> > +# use up 10 blocks of project quota
> > +$XFS_IO_PROG -f -c "pwrite -S 0x99 0 $((bsize * 10))" -c fsync $SCRATCH_MNT/dir/a >> $seqres.full
> > +
> > +# check statfs reporting
> > +fs_blocks=$(blocks $SCRATCH_MNT)
> > +dir_blocks=$(blocks $SCRATCH_MNT/dir)
> > +
> > +_within_tolerance "root blocks3" $fs_blocks $orig_blocks 1% -v
> > +_within_tolerance "dir blocks3" $dir_blocks $pquot_limit 1% -v
> > +
> > +fs_bavail=$(bavail $SCRATCH_MNT)
> > +dir_bavail=$(bavail $SCRATCH_MNT/dir)
> > +
> > +_within_tolerance "root bavail3" $fs_bavail $rem_free 1% -v
> > +_within_tolerance "dir bavail3" $dir_bavail $((rem_free - 10)) 1% -v
> > +
> > +# final state diagnostics
> > +$XFS_IO_PROG -c 'statfs' $SCRATCH_MNT $SCRATCH_MNT/dir | grep statfs >> $seqres.full
> > +repquota -P $SCRATCH_DEV >> $seqres.full
> > +df $SCRATCH_MNT >> $seqres.full
> > +ls -laR $SCRATCH_MNT/ >> $seqres.full
> > +
> > +# success, all done
> > +status=0
> > +exit
> > diff --git a/tests/generic/1955.out b/tests/generic/1955.out
> > new file mode 100644
> > index 00000000000000..3601010962193e
> > --- /dev/null
> > +++ b/tests/generic/1955.out
> > @@ -0,0 +1,13 @@
> > +QA output created by 1955
> > +root blocks1 is in range
> > +dir blocks1 is in range
> > +root bavail1 is in range
> > +dir bavail1 is in range
> > +root blocks2 is in range
> > +dir blocks2 is in range
> > +root bavail2 is in range
> > +dir bavail2 is in range
> > +root blocks3 is in range
> > +dir blocks3 is in range
> > +root bavail3 is in range
> > +dir bavail3 is in range
> >
>
>
prev parent reply other threads:[~2025-03-04 18:27 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <173992590656.4080455.15086949489894120802.stgit@frogsfrogsfrogs>
[not found] ` <173992590675.4080455.17713454161928793525.stgit@frogsfrogsfrogs>
2025-03-02 16:26 ` [PATCH 1/1] common: test statfs reporting with project quota Zorro Lang
2025-03-04 18:26 ` Darrick J. Wong [this message]
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=20250304182659.GC2803740@frogsfrogsfrogs \
--to=djwong@kernel.org \
--cc=fstests@vger.kernel.org \
--cc=linux-ext4@vger.kernel.org \
--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