Received: by 2002:a05:6a10:5bc5:0:0:0:0 with SMTP id os5csp1896415pxb; Thu, 4 Nov 2021 10:17:45 -0700 (PDT) X-Google-Smtp-Source: ABdhPJw7HWSxNYnORvYF1hKf6eqXC7yQzmRewA5/L+egHPwOPxUr2/5NLjqgwrvkOZWHAyIVRnnM X-Received: by 2002:a17:907:1b11:: with SMTP id mp17mr65235925ejc.275.1636046264902; Thu, 04 Nov 2021 10:17:44 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1636046264; cv=none; d=google.com; s=arc-20160816; b=Z9d4BCmpHjCfEeBhuqDSYBTCpGfEPvwRYGViQjZQ9Vev6JEsUEEjjGPwaOpaIvdpCT VI+HJoRfWd/aueoqmDYu2S5U+J9Z5NkKd1l/i6/bLLqTpUywY9f1KDa2n4y6vKGT80VH tZs2fttbtdFCvMTV1u3aRWHgVnX+VZ4VZbuFpDL9gmApJK1YiNvAsXerQ+OunTCPr93w xLJssJqrBjrNJE8BKEIVu5c61GYKoygpJfGC72xFZG6s0HEKF1tCH3XD1EOH9skJQo0L HxCKWnjWvwowowU3zX1shyn3z28/FPkqiUbjHdIp+31xMvCYHpVHVBls5xeEMFNaeWfV yb8Q== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:in-reply-to:content-disposition:mime-version :references:message-id:subject:cc:to:from:date:dkim-signature; bh=b4MlGzEfoAMgoWXy9qoblogWY2iSAL4p268g3W6zUGY=; b=AyEX1T4cO/+ZLFmfx08b3aAuWGaBr5s1ddf+a+Qy/0Y03GMdYz1BA4+cYpBpo3msrT 5r+L23vA8Id6sqmsjpOIa1qraUAmR4Rog7CeKJPX6XsbcQGlDj+7RdxLd8AqwDDnG02F +SC3pFO83evYXqaGyW3Ea5OqmOqTR6FH9CbC5WkBluXAPkq2++SY7m4qwNFhKlRp2OUE EQ5EuivchnuCWUoqKg/qnaNdDhp+2o3/4EV+ZCzt07e9vqxI6/W7FuCnoc9fmb4O69bV BDNSM1MoOL6zfGp+wIzSSPNsFilqtTNHcu2itSxCR8v5ddBIyd+ycIujmA21pKXltYsE hkJA== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@kernel.org header.s=k20201202 header.b=qjFi5McJ; spf=pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) smtp.mailfrom=linux-ext4-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Return-Path: Received: from vger.kernel.org (vger.kernel.org. [23.128.96.18]) by mx.google.com with ESMTP id sh14si10632888ejc.287.2021.11.04.10.16.59; Thu, 04 Nov 2021 10:17:44 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18; Authentication-Results: mx.google.com; dkim=pass header.i=@kernel.org header.s=k20201202 header.b=qjFi5McJ; spf=pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) smtp.mailfrom=linux-ext4-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233286AbhKDRTc (ORCPT + 99 others); Thu, 4 Nov 2021 13:19:32 -0400 Received: from mail.kernel.org ([198.145.29.99]:44704 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233283AbhKDRTc (ORCPT ); Thu, 4 Nov 2021 13:19:32 -0400 Received: by mail.kernel.org (Postfix) with ESMTPSA id DA10660720; Thu, 4 Nov 2021 17:16:53 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1636046214; bh=Me4o/mwSujSCm2YZx2jgtiv5rOj8AHHpNofSkXJ5z1Q=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=qjFi5McJc4G1PibhdjbYDilkYndFqFnFSUD+ufg0Myv+b4lfClWIFaJMpyuQnINHG 3XFAvQfxEtCNqVyht6pqQSZUFXvGGzUxAjKE4kWeTPitlQR1b58Y5LBg0hyEUqoSae rBMgYvzIhJV/qzUxjAWDlkjk4Ss3IOFJGRVGTHsBr8zx7BO/lIFDJSIyT3DdHasF3o fKgvS2ScamWXUTQBgNVhTPPT/LghWiTaSkW5eTabYz9Vp9WjzKpCxhYiZuiR/djJ3D 9fZXcq3PawbBHCeDbj2wLtCwxAfnA5MaFKv9idnDkzliQngtC0WTOZt8unHyIDFDDO kVqxQbXFscBFg== Date: Thu, 4 Nov 2021 10:16:53 -0700 From: "Darrick J. Wong" To: Lukas Czerner Cc: fstests@vger.kernel.org, linux-ext4@vger.kernel.org, tytso@mit.edu, guan@eryu.me Subject: Re: [PATCH v5] ext4: add test for all ext4/ext3/ext2 mount options Message-ID: <20211104171653.GY24282@magnolia> References: <20211102105911.5790-1-lczerner@redhat.com> <20211104102450.22965-1-lczerner@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20211104102450.22965-1-lczerner@redhat.com> Precedence: bulk List-ID: X-Mailing-List: linux-ext4@vger.kernel.org On Thu, Nov 04, 2021 at 11:24:50AM +0100, Lukas Czerner wrote: > Add test to validate that all the ext4, ext3 and ext2 are properly > recognized, validate and applied to avoid regressions as ext4 moves to > the new mount API. > > Signed-off-by: Lukas Czerner > --- > V2: Move minimum kernel version requirement up to 5.12 > Rewrite kernel_gte() to work correctly > V3: Rename functions to not start with _ > V4: Check kernel config file to make sure the test works with > various kernel configuration > V5: Check /proc/config.gz for kernel config location first > > tests/ext4/053 | 650 +++++++++++++++++++++++++++++++++++++++++++++ > tests/ext4/053.out | 2 + > 2 files changed, 652 insertions(+) > create mode 100755 tests/ext4/053 > create mode 100644 tests/ext4/053.out > > diff --git a/tests/ext4/053 b/tests/ext4/053 > new file mode 100755 > index 00000000..c4447623 > --- /dev/null > +++ b/tests/ext4/053 > @@ -0,0 +1,650 @@ > +#! /bin/bash > +# SPDX-License-Identifier: GPL-2.0 > +# Copyright (c) 2021 Red Hat, Inc., Lukas Czerner . > +# > +# FS QA Test 053 > +# > +# Sanity check of ext4 mount options > +# > +. ./common/preamble > +_begin_fstest auto mount > + > +seq=`basename $0` > +seqres=$RESULT_DIR/$seq > + > +here=`pwd` > +tmp=/tmp/$$ > +status=1 # failure is the default! > +trap "_cleanup; exit \$status" 0 1 2 3 15 > + > +_cleanup() > +{ > + cd / > + $UMOUNT_PROG $SCRATCH_MNT > /dev/null 2>&1 > + if [ -n "$LOOP_LOGDEV" ];then > + _destroy_loop_device $LOOP_LOGDEV 2>/dev/null > + fi > + rm -f $tmp.* > +} > + > +_has_kernel_config() > +{ > + option=$1 > + config="/proc/config.gz" > + if [ -f $config ]; then > + gzip -cd $config | grep -qE "^${option}=[my]" > + return > + fi > + config="/lib/modules/$(uname -r)/build/.config" > + grep -qE "^${option}=[my]" $config This probably ought to check /boot/config-$(uname -r) also, since that file seems to exist on RH and Debian. (The rest of the logic looks reasonable, but I've been away from ext4 for so long that I don't really have enough of a clue anymore... :( ) --D > +} > + > +_require_kernel_config() > +{ > + _has_kernel_config $1 || _notrun "Installed kernel not built with $1" > +} > + > +# get standard environment, filters and checks > +. ./common/rc > +. ./common/filter > +. ./common/quota > + > +# remove previous $seqres.full before test > +rm -f $seqres.full > + > +echo "Silence is golden." > + > +SIZE=$((1024 * 1024)) # 1GB in KB > +LOGSIZE=$((10 *1024)) # 10MB in KB > + > +_supported_fs ext2 ext3 ext4 > +_require_scratch_size $SIZE > +_require_quota > +_require_loop > +_require_command "$TUNE2FS_PROG" tune2fs > +MKE2FS_PROG=$(type -P mke2fs) > +_require_command "$MKE2FS_PROG" mke2fs > +_require_kernel_config CONFIG_QFMT_V2 > + > +LOG="" > +print_log() { > + LOG="$LOG $@" > +} > + > +KERNEL_VERSION=`uname -r | cut -d'.' -f1,2` > +KERNEL_MAJ=${KERNEL_VERSION%.*} > +KERNEL_MIN=${KERNEL_VERSION#*.} > + > +kernel_gte() { > + major=${1%.*} > + minor=${1#*.} > + > + if [ $KERNEL_MAJ -gt $major ]; then > + return 0 > + elif [[ $KERNEL_MAJ -eq $major && $KERNEL_MIN -ge $minor ]]; then > + return 0 > + fi > + return 1 > +} > + > + > +# The aim here is to keep the mount options behaviour stable going forward > +# so there is not much point in testing older kernels. > +kernel_gte 5.12 || _notrun "This test is only relevant for kernel versions 5.12 and higher" > + > +IGNORED="remount,defaults,ignored,removed" > +CHECK_MINFO="lazytime,noatime,nodiratime,noexec,nosuid,ro" > +ERR=0 > + > +test_mnt() { > + findmnt -n $SCRATCH_DEV > /dev/null 2>&1 > + [ $? -ne 0 ] && return $? > + > + if [ $# -eq 1 ]; then > + OPTS=$1 > + elif [ $# -eq 2 ]; then > + OPTS=$2 > + else > + return 0 > + fi > + > + print_log "checking \"$OPTS\" " > + # test options in /proc/fs/ext4/dev/options > + ( > + ret=0 > + IFS=',' > + for option in $OPTS; do > + if echo $IGNORED | grep -w $option; then > + continue > + fi > + > + [ $option = "noload" ] && option="norecovery" > + > + if [[ $option = ^* ]]; then > + expected=1 > + else > + expected=0 > + fi > + option=${option#^} > + > + if echo $CHECK_MINFO | grep -w $option; then > + findmnt -n -o OPTIONS $SCRATCH_DEV | grep $option > + ret=$? > + else > + grep $option /proc/fs/ext4/$(_short_dev $SCRATCH_DEV)/options > + ret=$? > + fi > + > + if [ $ret -ne $expected ]; then > + exit 1 > + fi > + done > + ) > /dev/null 2>&1 > + return $? > +} > + > +fail() { > + print_log " FAILED" > + ERR=$((ERR+1)) > + echo $LOG | tee -a $seqres.full > + LOG="" > +} > + > +ok() { > + print_log " OK" > + echo $LOG >> $seqres.full > + LOG="" > +} > + > +simple_mount() { > + _mount $* >> $seqres.full 2>&1 > +} > + > +# $1 - can hold -n option, if it does argumetns are shifted > +# $1 - options to test > +# $2 - if provided it's the option string to check for > +do_mnt() { > + device=$SCRATCH_DEV > + # If -n argument is provided do not specify $SCRATCH_DEV > + # usefull for remount > + if [ "$1" == "-n" ]; then > + unset device > + shift > + fi > + > + if [ -z "$1" ]; then > + simple_mount $device $SCRATCH_MNT > + ret=$? > + else > + simple_mount -o $1 $device $SCRATCH_MNT > + ret=$? > + fi > + if [ $ret -eq 0 ]; then > + test_mnt $1 $2 > + ret=$? > + [ $ret -ne 0 ] && print_log "(not found)" > + else > + print_log "(failed mount)" > + fi > + > + return $ret > +} > + > +not_mnt() { > + # We don't need -t not not_ variant > + if [ "$1" == "-t" ]; then > + shift > + fi > + > + print_log "SHOULD FAIL mounting $fstype \"$1\" " > + do_mnt $@ > + if [ $? -eq 0 ]; then > + fail > + else > + ok > + fi > + $UMOUNT_PROG $SCRATCH_MNT 2> /dev/null > +} > + > +mnt_only() { > + print_log "mounting $fstype \"$1\" " > + do_mnt $@ > + if [ $? -ne 0 ]; then > + fail > + else > + ok > + fi > +} > + > +mnt() { > + # Do we need to run the tune2fs mount option test ? > + t2fs=0 > + if [ "$1" == "-t" ]; then > + t2fs=1 > + shift > + fi > + > + mnt_only $* > + $UMOUNT_PROG $SCRATCH_MNT 2> /dev/null > + > + [ "$t2fs" -eq 0 ] && return > + > + op_set=$1 > + if [ $# -eq 1 ]; then > + check=$1 > + elif [ $# -eq 2 ]; then > + check=$2 > + else > + return 0 > + fi > + > + # some options need translation for tune2fs > + op_set=$(echo $op_set | sed -e 's/data=journal/journal_data/' \ > + -e 's/data=ordered/journal_data_ordered/' \ > + -e 's/data=writeback/journal_data_writeback/') > + $TUNE2FS_PROG -o $op_set $SCRATCH_DEV > /dev/null 2>&1 > + mnt_only "defaults" $check > + $UMOUNT_PROG $SCRATCH_MNT 2> /dev/null > + if [ "$op_set" = ^* ]; then > + op_set=${op_set#^} > + else > + op_set="^${op_set}" > + fi > + $TUNE2FS_PROG -o $op_set $SCRATCH_DEV > /dev/null 2>&1 > +} > + > +# $1 - options to mount with > +# $2 - options to remount with > +remount() { > + # First do this specifying both dev and mnt > + print_log "mounting $fstype \"$1\" " > + do_mnt $1 > + [ $? -ne 0 ] && fail && return > + print_log "remounting \"$2\" " > + do_mnt remount,$2 $3 > + if [ $? -ne 0 ]; then > + fail > + $UMOUNT_PROG $SCRATCH_MNT 2> /dev/null > + return > + else > + ok > + fi > + $UMOUNT_PROG $SCRATCH_MNT 2> /dev/null > + > + # Now just specify mnt > + print_log "mounting $fstype \"$1\" " > + do_mnt $1 > + [ $? -ne 0 ] && fail && return > + print_log "remounting (MNT ONLY) \"$2\" " > + do_mnt -n remount,$2 $3 > + if [ $? -ne 0 ]; then > + fail > + else > + ok > + fi > + > + $UMOUNT_PROG $SCRATCH_MNT 2> /dev/null > +} > + > +# $1 - options to mount with, or -r argument > +# $2 - options to remount with > +not_remount_noumount() { > + remount_only=0 > + # If -r is specified we're going to do remount only > + if [ "$1" == "-r" ]; then > + remount_only=1 > + # Dont need shift since first argument would > + # have been consumed by mount anyway > + fi > + > + if [ $remount_only -eq 0 ]; then > + print_log "mounting $fstype \"$1\" " > + do_mnt $1 > + [ $? -ne 0 ] && fail && return > + fi > + print_log "SHOULD FAIL remounting $fstype \"$2\" " > + do_mnt remount,$2 $3 > + if [ $? -eq 0 ]; then > + fail > + else > + ok > + fi > + > + # Now just specify mnt > + print_log "SHOULD FAIL remounting $fstype (MNT ONLY) \"$2\" " > + do_mnt -n remount,$2 $3 > + if [ $? -eq 0 ]; then > + fail > + else > + ok > + fi > +} > + > +not_remount() { > + not_remount_noumount $* > + $UMOUNT_PROG $SCRATCH_MNT 2> /dev/null > +} > + > + > +do_mkfs() { > + $MKE2FS_PROG -T $fstype -Fq $* >> $seqres.full 2>&1 || > + _fail "mkfs failed - $MKFS_EXT4_PROG -Fq $* $SCRATCH_DEV" > +} > + > +not_ext2() { > + if [[ $fstype == "ext2" ]]; then > + not_$* > + else > + $* > + fi > +} > + > +only_ext4() { > + if [[ $fstype == "ext4" ]]; then > + $* > + else > + not_$* > + fi > +} > + > +# Create logdev for external journal > +LOOP_IMG=$tmp.logdev > +truncate -s ${LOGSIZE}k $LOOP_IMG > +LOOP_LOGDEV=`_create_loop_device $LOOP_IMG` > +majmin=`stat -c "%t:%T" $LOOP_LOGDEV` > +LOGDEV_DEVNUM=`echo "${majmin%:*}*2^8 + ${majmin#*:}" | bc` > + > +# Test all the extN file system supported by ext4 driver > +fstype= > +for fstype in ext2 ext3 ext4; do > + > + $UMOUNT_PROG $SCRATCH_MNT 2> /dev/null > + $UMOUNT_PROG $SCRATCH_DEV 2> /dev/null > + > + do_mkfs $SCRATCH_DEV ${SIZE}k > + > + # do we have fstype support ? > + do_mnt > + if [ $? -ne 0 ]; then > + print_log "$fstype not supported. Skipping..." > + ok > + continue > + fi > + if [ ! -f /proc/fs/ext4/$(_short_dev $SCRATCH_DEV)/options ]; then > + print_log "$fstype not supported. Skipping..." > + ok > + continue > + fi > + > + $UMOUNT_PROG $SCRATCH_MNT 2> /dev/null > + > + not_mnt failme > + mnt > + mnt bsddf > + mnt minixdf > + mnt grpid > + mnt -t bsdgroups grpid > + mnt nogrpid > + mnt sysvgroups nogrpid > + mnt resgid=1001 > + mnt resuid=1001 > + mnt sb=131072 > + mnt errors=continue > + mnt errors=panic > + mnt errors=remount-ro > + mnt nouid32 > + mnt debug > + mnt oldalloc removed > + mnt orlov removed > + mnt -t user_xattr > + mnt nouser_xattr > + > + if _has_kernel_config CONFIG_EXT4_FS_POSIX_ACL; then > + mnt -t acl > + else > + mnt -t acl ^acl > + fi > + > + not_ext2 mnt noload norecovery > + mnt bh removed > + mnt nobh removed > + not_ext2 mnt commit=7 > + mnt min_batch_time=200 > + mnt max_batch_time=10000 > + only_ext4 mnt journal_checksum > + only_ext4 mnt nojournal_checksum > + only_ext4 mnt journal_async_commit,data=writeback > + mnt abort ignored > + not_ext2 mnt -t data=journal > + not_ext2 mnt -t data=ordered > + not_ext2 mnt -t data=writeback > + not_ext2 mnt data_err=abort > + not_ext2 mnt data_err=ignore ignored > + mnt usrjquota=aquota.user,jqfmt=vfsv0 > + not_mnt usrjquota=aquota.user > + mnt usrjquota= ignored > + mnt grpjquota=aquota.group,jqfmt=vfsv0 > + not_mnt grpjquota=aquota.group > + mnt grpjquota= ignored > + mnt jqfmt=vfsold > + mnt jqfmt=vfsv0 > + mnt jqfmt=vfsv1 > + mnt grpquota > + mnt quota > + mnt noquota > + mnt usrquota > + mnt grpquota > + mnt barrier > + mnt barrier=0 nobarrier > + mnt barrier=1 barrier > + mnt barrier=99 barrier > + mnt -t nobarrier > + mnt i_version > + mnt stripe=512 > + only_ext4 mnt delalloc > + only_ext4 mnt -t nodelalloc > + mnt warn_on_error > + mnt nowarn_on_error > + not_mnt debug_want_extra_isize=512 > + mnt debug_want_extra_isize=32 ignored > + mnt mblk_io_submit removed > + mnt nomblk_io_submit removed > + mnt -t block_validity > + mnt noblock_validity > + mnt inode_readahead_blks=16 > + not_ext2 mnt journal_ioprio=6 ignored > + mnt auto_da_alloc=0 noauto_da_alloc > + mnt auto_da_alloc=1 auto_da_alloc > + mnt auto_da_alloc=95 auto_da_alloc > + mnt auto_da_alloc > + mnt noauto_da_alloc > + only_ext4 mnt dioread_nolock > + only_ext4 mnt nodioread_nolock > + only_ext4 mnt dioread_lock nodioread_nolock > + mnt -t discard > + mnt nodiscard > + mnt init_itable=20 > + mnt init_itable > + mnt init_itable=0 > + mnt noinit_itable > + mnt max_dir_size_kb=4096 > + > + if _has_kernel_config CONFIG_FS_ENCRYPTION; then > + mnt test_dummy_encryption > + mnt test_dummy_encryption=v1 > + mnt test_dummy_encryption=v2 > + not_mnt test_dummy_encryption=v3 > + not_mnt test_dummy_encryption= > + else > + mnt test_dummy_encryption ^test_dummy_encryption > + mnt test_dummy_encryption=v1 ^test_dummy_encryption=v1 > + mnt test_dummy_encryption=v2 ^test_dummy_encryption=v2 > + not_mnt test_dummy_encryption=v3 > + not_mnt test_dummy_encryption= > + fi > + > + if _has_kernel_config CONFIG_FS_ENCRYPTION_INLINE_CRYPT; then > + mnt inlinecrypt > + else > + mnt inlinecrypt ^inlinecrypt > + fi > + > + mnt prefetch_block_bitmaps removed > + mnt no_prefetch_block_bitmaps > + # We don't currently have a way to know that the option has been > + # applied, so comment it out for now. This should be fixed in the > + # future. > + #mnt mb_optimize_scan=0 > + #mnt mb_optimize_scan=1 > + #not_mnt mb_optimize_scan=9 > + #not_mnt mb_optimize_scan= > + mnt nombcache > + mnt no_mbcache nombcache > + mnt check=none removed > + mnt nocheck removed > + mnt reservation removed > + mnt noreservation removed > + mnt journal=20 ignored > + not_mnt nonsenseoption > + not_mnt nonsenseoption=value > + > + # generic mount options > + mnt lazytime > + mnt nolazytime ^lazytime > + mnt noatime > + mnt nodiratime > + mnt noexec > + mnt nosuid > + mnt ro > + > + # generic remount check > + remount barrier nobarrier > + remount nobarrier barrier > + remount discard nodiscard > + remount nodiscard discard > + > + # dax mount options > + simple_mount -o dax=always $SCRATCH_DEV $SCRATCH_MNT > /dev/null 2>&1 > + if [ $? -eq 0 ]; then > + $UMOUNT_PROG $SCRATCH_MNT 2> /dev/null > + mnt dax > + mnt dax=always > + mnt dax=never > + mnt dax=inode > + > + not_remount lazytime dax > + not_remount dax=always dax=never > + > + if [[ $fstype != "ext2" ]]; then > + not_remount data=journal dax > + not_remount data=journal dax=always > + not_remount data=journal dax=never > + not_remount data=journal dax=inode > + fi > + fi > + > + # Quota remount check > + remount grpquota usrquota > + remount usrquota quota > + remount usrquota usrjquota=q.u,jqfmt=vfsv0 > + remount grpquota grpjquota=q.g,jqfmt=vfsv0 > + > + not_remount usrquota grpjquota=q.g,jqfmt=vfsv0 > + not_remount grpquota usrjquota=q.u,jqfmt=vfsv0 > + > + remount quota usrjquota=q.u,jqfmt=vfsv0 > + not_remount quota grpjquota=q.g,jqfmt=vfsv0 > + > + remount usrjquota=q.u,jqfmt=vfsv0 grpjquota=q.g > + not_remount usrjquota=q.u,jqfmt=vfsv0 usrjquota=q.ua > + not_remount grpjquota=q.g,jqfmt=vfsv0 grpjquota=q.ga > + > + remount usrjquota=q.u,jqfmt=vfsv0 usrquota usrjquota=q.u,jqfmt=vfsv0 > + remount grpjquota=q.g,jqfmt=vfsv0 grpquota grpjquota=q.g,jqfmt=vfsv0 > + not_remount usrjquota=q.u,jqfmt=vfsv0 grpquota > + not_remount grpjquota=q.g,jqfmt=vfsv0 usrquota > + > + remount grpjquota=q.g,jqfmt=vfsv0 grpjquota= ^grpjquota= > + remount usrjquota=q.u,jqfmt=vfsv0 usrjquota= ^usrjquota= > + remount grpjquota=q.g,usrjquota=q.u,jqfmt=vfsv0 grpjquota=,usrjquota= ^grpjquota=,^usrjquota= > + > + remount jqfmt=vfsv0 grpjquota=q.g > + remount jqfmt=vfsv0 usrjquota=q.u > + > + if [[ $fstype != "ext2" ]]; then > + remount noload data=journal norecovery > + not_remount data=ordered data=journal > + not_remount data=journal data=writeback > + not_remount data=writeback data=ordered > + fi > + > + do_mkfs -O journal_dev $LOOP_LOGDEV ${LOGSIZE}k > + do_mkfs -J device=$LOOP_LOGDEV $SCRATCH_DEV ${SIZE}k > + mnt defaults > + mnt journal_path=$LOOP_LOGDEV ignored > + mnt journal_dev=$LOGDEV_DEVNUM ignored > + not_mnt journal_path=${LOOP_LOGDEV}_nonexistent ignored > + not_mnt journal_dev=123456 ignored > + not_mnt journal_dev=999999999999999 ignored > + > + do_mkfs -E quotatype=prjquota $SCRATCH_DEV ${SIZE}k > + mnt prjquota > + > + # test clearing/changing journalled quota when enabled > + echo "== Testing active journalled quota" >> $seqres.full > + mnt_only prjquota,grpjquota=aquota.group,usrjquota=aquota.user,jqfmt=vfsv0 > + > + # Prepare and enable quota > + quotacheck -vugm $SCRATCH_MNT >> $seqres.full 2>&1 > + quotaon -vug $SCRATCH_MNT >> $seqres.full 2>&1 > + > + not_remount_noumount -r grpjquota= > + not_remount_noumount -r usrjquota=aaquota.user > + not_remount_noumount -r grpjquota=aaquota.group > + not_remount_noumount -r jqfmt=vfsv1 > + not_remount_noumount -r noquota > + mnt_only remount,usrquota,grpquota ^usrquota,^grpquota > + $UMOUNT_PROG $SCRATCH_MNT > /dev/null 2>&1 > + > + # test clearing/changing quota when enabled > + do_mkfs -E quotatype=^prjquota $SCRATCH_DEV ${SIZE}k > + not_mnt prjquota > + echo "== Testing active non-journalled quota" >> $seqres.full > + mnt_only grpquota,usrquota > + > + # Prepare and enable quota > + quotacheck -vugm $SCRATCH_MNT >> $seqres.full 2>&1 > + quotaon -vug $SCRATCH_MNT >> $seqres.full 2>&1 > + > + not_remount_noumount -r noquota > + not_remount_noumount -r usrjquota=aquota.user > + not_remount_noumount -r grpjquota=aquota.group > + not_remount_noumount -r jqfmt=vfsv1 > + mnt_only remount,grpjquota= grpquota,^grpjquota > + mnt_only remount,usrjquota= usrquota,^usrjquota > + mnt_only remount,usrquota,grpquota usrquota,grpquota > + quotaoff -f $SCRATCH_MNT >> $seqres.full 2>&1 > + mnt_only remount,noquota ^usrquota,^grpquota,quota > + $UMOUNT_PROG $SCRATCH_MNT > /dev/null 2>&1 > + > + # Quota feature > + echo "== Testing quota feature " >> $seqres.full > + do_mkfs -O quota -E quotatype=prjquota $SCRATCH_DEV ${SIZE}k > + mnt usrjquota=aquota.user,jqfmt=vfsv0 ^usrjquota= > + mnt grpjquota=aquota.user,jqfmt=vfsv0 ^grpjquota= > + mnt jqfmt=vfsv1 ^jqfmt= > + mnt prjquota > + mnt usrquota > + mnt grpquota > + not_remount defaults usrjquota=aquota.user > + not_remount defaults grpjquota=aquota.user > + not_remount defaults jqfmt=vfsv1 > + remount defaults grpjquota=,usrjquota= ignored > + > +done #for fstype in ext2 ext3 ext4; do > + > +$UMOUNT_PROG $SCRATCH_MNT > /dev/null 2>&1 > +echo "$ERR errors encountered" >> $seqres.full > + > +status=$ERR > +exit > diff --git a/tests/ext4/053.out b/tests/ext4/053.out > new file mode 100644 > index 00000000..d58db7e4 > --- /dev/null > +++ b/tests/ext4/053.out > @@ -0,0 +1,2 @@ > +QA output created by 053 > +Silence is golden. > -- > 2.31.1 >