| 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
| ||
|
Message-ID: <20220119043701.GC13563@magnolia>
Date: Tue, 18 Jan 2022 20:37:01 -0800
From: "Darrick J. Wong" <djwong@...nel.org>
To: Dan Carpenter <dan.carpenter@...cle.com>
Cc: kbuild@...ts.01.org, lkp@...el.com, kbuild-all@...ts.01.org,
"Darrick J. Wong" <darrick.wong@...cle.com>,
linux-kernel@...r.kernel.org
Subject: Re: [djwong-xfs:vectorized-scrub 99/334]
fs/xfs/scrub/fscounters.c:198 xchk_fscounters_freeze() warn: inconsistent
returns '&sb->s_umount'.
On Wed, Jan 19, 2022 at 06:58:29AM +0300, Dan Carpenter wrote:
> tree: https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux.git vectorized-scrub
> head: 8427da8e62fbcf9a04e5b2663fe60b97d6911417
> commit: 8dd594d12f08acc6c6fa388b2cae3e270bf8effc [99/334] xfs: stabilize fs summary counters for online fsck
> config: ia64-randconfig-m031-20220116 (https://download.01.org/0day-ci/archive/20220117/202201170928.GcIhOWMI-lkp@intel.com/config)
> compiler: ia64-linux-gcc (GCC) 11.2.0
>
> If you fix the issue, kindly add following tag as appropriate
> Reported-by: kernel test robot <lkp@...el.com>
> Reported-by: Dan Carpenter <dan.carpenter@...cle.com>
>
> smatch warnings:
> fs/xfs/scrub/fscounters.c:198 xchk_fscounters_freeze() warn: inconsistent returns '&sb->s_umount'.
>
> vim +198 fs/xfs/scrub/fscounters.c
>
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 125 STATIC int
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 126 xchk_fscounters_freeze(
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 127 struct xfs_scrub *sc)
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 128 {
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 129 struct xchk_fscounters *fsc = sc->buf;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 130 struct xfs_mount *mp = sc->mp;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 131 struct super_block *sb = mp->m_super;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 132 int level;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 133 int error = 0;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 134
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 135 if (sc->flags & XCHK_HAVE_FREEZE_PROT) {
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 136 sc->flags &= ~XCHK_HAVE_FREEZE_PROT;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 137 mnt_drop_write_file(sc->file);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 138 }
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 139
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 140 /* Wait until we're ready to freeze or give up. */
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 141 down_write(&sb->s_umount);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 142 while (sb->s_writers.frozen != SB_UNFROZEN) {
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 143 up_write(&sb->s_umount);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 144
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 145 if (xchk_should_terminate(sc, &error))
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 146 return error;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 147
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 148 delay(HZ / 10);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 149 down_write(&sb->s_umount);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 150 }
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 151
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 152 if (sb_rdonly(sb)) {
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 153 sb->s_writers.frozen = SB_FREEZE_EXCLUSIVE;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 154 goto done;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 155 }
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 156
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 157 sb->s_writers.frozen = SB_FREEZE_WRITE;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 158 /* Release s_umount to preserve sb_start_write -> s_umount ordering */
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 159 up_write(&sb->s_umount);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 160 percpu_down_write(sb->s_writers.rw_sem + SB_FREEZE_WRITE - 1);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 161 down_write(&sb->s_umount);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 162
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 163 /* Now we go and block page faults... */
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 164 sb->s_writers.frozen = SB_FREEZE_PAGEFAULT;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 165 percpu_down_write(sb->s_writers.rw_sem + SB_FREEZE_PAGEFAULT - 1);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 166
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 167 /*
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 168 * All writers are done so after syncing there won't be dirty data.
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 169 * Let xfs_fs_sync_fs flush dirty data so the VFS won't start writeback
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 170 * and to disable the background gc workers.
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 171 */
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 172 error = sync_filesystem(sb);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 173 if (error) {
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 174 sb->s_writers.frozen = SB_UNFROZEN;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 175 for (level = SB_FREEZE_PAGEFAULT; level >= 0; level--)
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 176 percpu_up_write(sb->s_writers.rw_sem + level);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 177 wake_up(&sb->s_writers.wait_unfrozen);
>
> Smatch wanted an up_write(&sb->s_umount); but this looks intentional?
Nope, that's a bug. Thanks for catching that!
--D
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 178 return error;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 179 }
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 180
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 181 /* Now wait for internal filesystem counter */
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 182 sb->s_writers.frozen = SB_FREEZE_FS;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 183 percpu_down_write(sb->s_writers.rw_sem + SB_FREEZE_FS - 1);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 184
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 185 /*
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 186 * We do not need to quiesce the log to check the summary counters, so
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 187 * skip the call to xfs_fs_freeze here. To prevent anyone else from
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 188 * unfreezing us, set the VFS freeze level to one higher than
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 189 * FREEZE_COMPLETE.
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 190 */
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 191 sb->s_writers.frozen = SB_FREEZE_EXCLUSIVE;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 192 for (level = SB_FREEZE_LEVELS - 1; level >= 0; level--)
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 193 percpu_rwsem_release(sb->s_writers.rw_sem + level, 0,
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 194 _THIS_IP_);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 195 done:
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 196 fsc->frozen = true;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 197 up_write(&sb->s_umount);
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 @198 return 0;
> 8dd594d12f08acc Darrick J. Wong 2022-01-06 199 }
>
> ---
> 0-DAY CI Kernel Test Service, Intel Corporation
> https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
>
Powered by blists - more mailing lists