From nobody Sat Dec 27 01:29:20 2025 Received: from szxga05-in.huawei.com (szxga05-in.huawei.com [45.249.212.191]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id D067E10E4 for ; Thu, 28 Dec 2023 01:33:42 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=huawei.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=huawei.com Received: from mail.maildlp.com (unknown [172.19.163.17]) by szxga05-in.huawei.com (SkyGuard) with ESMTP id 4T0rXQ6MYxz1FFnD; Thu, 28 Dec 2023 09:29:42 +0800 (CST) Received: from kwepemm000013.china.huawei.com (unknown [7.193.23.81]) by mail.maildlp.com (Postfix) with ESMTPS id C994B1A0173; Thu, 28 Dec 2023 09:33:34 +0800 (CST) Received: from huawei.com (10.175.127.227) by kwepemm000013.china.huawei.com (7.193.23.81) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2507.35; Thu, 28 Dec 2023 09:33:34 +0800 From: Zhihao Cheng To: , , , , CC: , Subject: [PATCH mtd-utils 03/11] tests: ubifs_repair: Add cycle mount+repair test Date: Thu, 28 Dec 2023 09:36:31 +0800 Message-ID: <20231228013639.2827205-4-chengzhihao1@huawei.com> X-Mailer: git-send-email 2.31.1 In-Reply-To: <20231228013639.2827205-1-chengzhihao1@huawei.com> References: <20231228013639.2827205-1-chengzhihao1@huawei.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-ClientProxiedBy: dggems701-chm.china.huawei.com (10.3.19.178) To kwepemm000013.china.huawei.com (7.193.23.81) Content-Type: text/plain; charset="utf-8" Do fsstress and repair ubifs image, make sure all files(and their data) are not lost after repairing. This testcase mainly checks whether ubifs_repair could corrupt filesystem content in common case. Signed-off-by: Zhihao Cheng --- .gitignore | 1 + configure.ac | 3 +- tests/ubifs_repair-tests/Makemodule.am | 3 +- .../tests/cycle_mount_repair_check.sh.in | 177 +++++++++++++++++= ++++ 4 files changed, 182 insertions(+), 2 deletions(-) create mode 100755 tests/ubifs_repair-tests/tests/cycle_mount_repair_check= .sh.in diff --git a/.gitignore b/.gitignore index 7613349..1a5fe7e 100644 --- a/.gitignore +++ b/.gitignore @@ -114,6 +114,7 @@ tests/ubi-tests/runubitests.sh tests/ubi-tests/ubi-stress-test.sh tests/ubifs_repair-tests/lib/common.sh tests/ubifs_repair-tests/tests/authentication_refuse.sh +tests/ubifs_repair-tests/tests/cycle_mount_repair_check.sh =20 # # Files generated by autotools diff --git a/configure.ac b/configure.ac index a42a55d..73a3853 100644 --- a/configure.ac +++ b/configure.ac @@ -356,6 +356,7 @@ AC_CONFIG_FILES([tests/fs-tests/fs_help_all.sh tests/ubi-tests/runubitests.sh tests/ubi-tests/ubi-stress-test.sh tests/ubifs_repair-tests/lib/common.sh - tests/ubifs_repair-tests/tests/authentication_refuse.sh]) + tests/ubifs_repair-tests/tests/authentication_refuse.sh + tests/ubifs_repair-tests/tests/cycle_mount_repair_check.sh]) =20 AC_OUTPUT([Makefile]) diff --git a/tests/ubifs_repair-tests/Makemodule.am b/tests/ubifs_repair-te= sts/Makemodule.am index c0a6ea1..c2556f5 100644 --- a/tests/ubifs_repair-tests/Makemodule.am +++ b/tests/ubifs_repair-tests/Makemodule.am @@ -1,3 +1,4 @@ test_SCRIPTS +=3D \ tests/ubifs_repair-tests/lib/common.sh \ - tests/ubifs_repair-tests/tests/authentication_refuse.sh + tests/ubifs_repair-tests/tests/authentication_refuse.sh \ + tests/ubifs_repair-tests/tests/cycle_mount_repair_check.sh diff --git a/tests/ubifs_repair-tests/tests/cycle_mount_repair_check.sh.in = b/tests/ubifs_repair-tests/tests/cycle_mount_repair_check.sh.in new file mode 100755 index 0000000..1f5db3f --- /dev/null +++ b/tests/ubifs_repair-tests/tests/cycle_mount_repair_check.sh.in @@ -0,0 +1,177 @@ +#!/bin/sh +# Copyright (c), 2023-2024, Huawei Technologies Co, Ltd. +# Author: Zhihao Cheng +# +# Test Description: +# Do many cycles of mount/fsstress/umount/repair/mount, check whether file= system +# content before repairing and after repairing are consistent. +# Running time: 16h + +TESTBINDIR=3D@TESTBINDIR@ +source $TESTBINDIR/common.sh + +# No need to test on large size flash, it will take very long time iterati= ng +# all files in large filesystem. +ID=3D"0xec,0xa1,0x00,0x15" # 128M 128KB 2KB 512-sub-page + +function run_test() +{ + mode=3D$1 + encryption=3D$2 + modprobe nandsim id_bytes=3D$ID + mtdnum=3D"$(find_mtd_device "$nandsim_patt")" + flash_eraseall /dev/mtd$mtdnum + + dmesg -c > /dev/null + + modprobe ubi mtd=3D"$mtdnum,2048,0,0,1" fm_autoconvert || fatal "modprobe= ubi fail" + ubimkvol -N vol_test -m -n 0 /dev/ubi$UBI_NUM || fatal "mkvol fail" + modprobe ubifs || fatal "modprobe ubifs fail" + + echo "Do cycle mount+umount+repair+check_fs_content test ($encryption,$mo= de)" + + if [[ "$encryption" =3D=3D "encrypted" ]]; then + encryption_gen_key + fi + + round=3D0 + while [[ $round -lt 20 ]] + do + echo "---------------------- ROUND $round ----------------------" + let round=3D$round+1 + + mount_ubifs $DEV $MNT || fatal "mount ubifs fail" + if [[ "$encryption" =3D=3D "encrypted" ]]; then + encryption_set_key $MNT + fi + + per=3D`df -Th | grep ubifs | awk '{print $6}'`; + if [[ ${per%?} -gt 95 ]]; then + # Used > 95% + echo "Clean files" + rm -rf $MNT/* + check_err_msg + fi + + # Truncation and writing ops cannot coexist, because ubifs_repair could = recover truncated data nodes which leads different md5 results before repai= ring and after repairing. For example: + # 1. write(file, 8K) + # 2. truncate(file, 0) # data block 0,1 are truncated + # 3. lseek(file, 4K) + # 4. write(file, 4K) # data block 1' is created + # 5. truncate(file, 8000) # data block 1'' is modified + # It is expected that only data block 1'' is found after repairing, but = ubifs_repair will try best to recover all nodes(ubifs_repair cannot utilize= TNC information), so data block 0 and data block 1'' are found. + if [[ "$mode" =3D=3D "notruncate" ]] + then + fsstress -d $MNT -l0 -p4 -n10000 -f truncate=3D0 & + elif [[ "$mode" =3D=3D "nowrite" ]] + then + fsstress -d $MNT -l0 -p4 -n10000 -f dwrite=3D0 -f awrite=3D0 -f write= =3D0 -f writev=3D0 -f mwrite=3D0 -f uring_write=3D0 -f copyrange=3D0 -f ded= uperange=3D0 -f clonerange=3D0 -f splice=3D0 & + else + fsstress -d $MNT -l0 -p4 -n10000 & + fi + + sleep $((RANDOM % 30)) + + ps -e | grep -w fsstress > /dev/null 2>&1 + while [ $? -eq 0 ] + do + killall -9 fsstress > /dev/null 2>&1 + sleep 1 + ps -e | grep -w fsstress > /dev/null 2>&1 + done + + check_err_msg # Make sure new operations are okay after repairing UBIFS + sync + + # Record filesystem information + rm -f $TMP_FILE 2>/dev/null + if [[ "$mode" =3D=3D "all" ]] + then + read_dir $MNT "nomd5sum" + else + read_dir $MNT "md5sum" + fi + + while true + do + res=3D`mount | grep "$MNT"` + if [[ "$res" =3D=3D "" ]] + then + break; + fi + umount $MNT + sleep 0.1 + done + + echo 'format "UBIFS DBG repair" +pflmt' > /sys/kernel/debug/dynamic_debu= g/control + echo "$DEV" > /sys/kernel/debug/ubifs/repair_fs + res=3D$? + if [[ $res !=3D 0 ]] + then + fatal "repair fail $res" + fi + + # There are no powercut or corruption, error messages won't appear in UB= IFS repairing + check_err_msg + + check_memleak + + enable_chkfs + + mount_ubifs $DEV $MNT + res=3D$? + if [[ $res !=3D 0 ]] + then + fatal "mount fail $res" + fi + + if [[ "$encryption" =3D=3D "encrypted" ]]; then + encryption_set_key $MNT + fi + + du -sh $MNT > /dev/null # Make sure all files are accessible + ret=3D$? + if [[ $ret !=3D 0 ]]; then + fatal "Cannot access all files" + fi + check_err_msg + + # Check filesystem information + if [[ "$mode" =3D=3D "all" ]] + then + parse_dir "nomd5sum" + else + parse_dir "md5sum" + fi + rm -f $TMP_FILE 2>/dev/null + + umount $MNT + res=3D$? + if [[ $res !=3D 0 ]] + then + fatal "unmount fail $res" + fi + + check_err_msg + + disable_chkfs + done + + modprobe -r ubifs + modprobe -r ubi + modprobe -r nandsim +} + +check_fsstress +start_t=3D$(date +%s) +for encryption in "encrypted" "noencrypted"; do + # This situation is ignored, fsstress (comes from https://git.kernel.org/= pub/scm/fs/xfs/xfstests-dev.git) could invoke open(O_TRUNC), however there = is no 'O_TRUNC' passed into open syscall after looking through the code, so= we just skip this case for now. + #run_test "notruncate" $encryption + + run_test "nowrite" $encryption + run_test "all" $encryption +done +end_t=3D$(date +%s) +time_cost=3D$(( end_t - start_t )) +echo "Success, cost $time_cost seconds" +exit 0 --=20 2.13.6