From: Liu Bo <bo.li.liu@oracle.com>
To: "Tomasz Kłoczko" <kloczko.tomasz@gmail.com>
Cc: David Sterba <dsterba@suse.com>, Chris Mason <clm@fb.com>,
Linux fs Btrfs <linux-btrfs@vger.kernel.org>
Subject: Re: [GIT PULL] Btrfs fixes for 4.14-rc4
Date: Fri, 6 Oct 2017 16:23:39 -0700 [thread overview]
Message-ID: <20171006232339.GA19068@lim.localdomain> (raw)
In-Reply-To: <CABB28Cz_n8GgXDWcdC4+93mqAGgj+nxL7jF_Yn_J89shyHmPMw@mail.gmail.com>
On Fri, Oct 06, 2017 at 11:25:12PM +0100, Tomasz Kłoczko wrote:
> On rc3 is possible to observe warning about possible circular locking
> dependency which I've reported on btrfs list few days ago:
>
Thanks for the report, neither this nor the one you reported on rc2
looks like a deadlock to me.
- %mmap_sem is always 'current->mm->mmap_sem',
- fsync doesn't acquire %mmap_sem, does it?
Not sure why we have a dependency here, tend to be a false one.
thanks,
-liubo
> [ 101.326724] ======================================================
> [ 101.326728] WARNING: possible circular locking dependency detected
> [ 101.326734] 4.14.0-0.rc3.git1.1.fc28.x86_64 #1 Not tainted
> [ 101.326738] ------------------------------------------------------
> [ 101.326743] mysqld/1253 is trying to acquire lock:
> [ 101.326747] (&mm->mmap_sem){++++}, at: [<ffffffffbd269a9e>]
> get_user_pages_unlocked+0x5e/0x1b0
> [ 101.326771]
> [ 101.326775] (&ei->dio_sem){++++}, at: [<ffffffffc02e8bdf>]
> btrfs_direct_IO+0x39f/0x400 [btrfs]
> [ 101.326846]
> [ 101.326851]
> [ 101.326856]
> [ 101.326875] __lock_acquire+0x1107/0x11d0
> [ 101.326883] lock_acquire+0xa3/0x1f0
> [ 101.326892] down_write+0x51/0xc0
> [ 101.326949] btrfs_log_changed_extents+0x7e/0x6c0 [btrfs]
> [ 101.327000] btrfs_log_inode+0x9c1/0x11d0 [btrfs]
> [ 101.327049] btrfs_log_inode_parent+0x2df/0xad0 [btrfs]
> [ 101.327096] btrfs_log_dentry_safe+0x60/0x80 [btrfs]
> [ 101.327144] btrfs_sync_file+0x344/0x4f0 [btrfs]
> [ 101.327155] vfs_fsync_range+0x4b/0xb0
> [ 101.327162] do_fsync+0x3d/0x70
> [ 101.327170] SyS_fsync+0x10/0x20
> [ 101.327179] do_syscall_64+0x6c/0x1f0
> [ 101.327185] return_from_SYSCALL_64+0x0/0x7a
> [ 101.327188]
> [ 101.327204] __lock_acquire+0x1107/0x11d0
> [ 101.327212] lock_acquire+0xa3/0x1f0
> [ 101.327219] __mutex_lock+0x7f/0xa40
> [ 101.327226] mutex_lock_nested+0x1b/0x20
> [ 101.327272] btrfs_log_inode+0x159/0x11d0 [btrfs]
> [ 101.327317] btrfs_log_inode_parent+0x2df/0xad0 [btrfs]
> [ 101.327360] btrfs_log_dentry_safe+0x60/0x80 [btrfs]
> [ 101.327407] btrfs_sync_file+0x344/0x4f0 [btrfs]
> [ 101.327415] vfs_fsync_range+0x4b/0xb0
> [ 101.327422] do_fsync+0x3d/0x70
> [ 101.327429] SyS_fsync+0x10/0x20
> [ 101.327435] do_syscall_64+0x6c/0x1f0
> [ 101.327441] return_from_SYSCALL_64+0x0/0x7a
> [ 101.327444]
> [ 101.327463] __sb_start_write+0x12b/0x1a0
> [ 101.327508] start_transaction+0x368/0x4d0 [btrfs]
> [ 101.327549] btrfs_join_transaction+0x1d/0x20 [btrfs]
> [ 101.327589] delayed_ref_async_start+0x67/0xd0 [btrfs]
> [ 101.327637] btrfs_worker_helper+0x93/0x610 [btrfs]
> [ 101.327640]
> [ 101.327656] __lock_acquire+0x1107/0x11d0
> [ 101.327664] lock_acquire+0xa3/0x1f0
> [ 101.327671] wait_for_completion+0x62/0x1d0
> [ 101.327710] btrfs_async_run_delayed_refs+0x163/0x180 [btrfs]
> [ 101.327752] __btrfs_end_transaction+0x1f2/0x2e0 [btrfs]
> [ 101.327790] btrfs_end_transaction+0x10/0x20 [btrfs]
> [ 101.327832] btrfs_dirty_inode+0x71/0xd0 [btrfs]
> [ 101.327871] btrfs_update_time+0x81/0xc0 [btrfs]
> [ 101.327877] touch_atime+0xab/0xd0
> [ 101.327920] btrfs_file_mmap+0x44/0x60 [btrfs]
> [ 101.327927] mmap_region+0x3a3/0x5d0
> [ 101.327932] do_mmap+0x2b6/0x410
> [ 101.327938] vm_mmap_pgoff+0xcf/0x120
> [ 101.327943] SyS_mmap_pgoff+0x1e1/0x280
> [ 101.327949] SyS_mmap+0x1b/0x30
> [ 101.327955] entry_SYSCALL_64_fastpath+0x1f/0xbe
> [ 101.327958]
> [ 101.327974] check_prev_add+0x351/0x700
> [ 101.327981] __lock_acquire+0x1107/0x11d0
> [ 101.327989] lock_acquire+0xa3/0x1f0
> [ 101.327996] down_read+0x48/0xb0
> [ 101.328003] get_user_pages_unlocked+0x5e/0x1b0
> [ 101.328009] get_user_pages_fast+0x7a/0xc0
> [ 101.328018] iov_iter_get_pages+0xc9/0x300
> [ 101.328026] do_blockdev_direct_IO+0x192b/0x2940
> [ 101.328034] __blockdev_direct_IO+0x2e/0x30
> [ 101.328073] btrfs_direct_IO+0x171/0x400 [btrfs]
> [ 101.328080] generic_file_direct_write+0xa3/0x160
> [ 101.328123] btrfs_file_write_iter+0x2fb/0x610 [btrfs]
> [ 101.328129] aio_write+0x116/0x1a0
> [ 101.328134] do_io_submit+0x42d/0x940
> [ 101.328139] SyS_io_submit+0x10/0x20
> [ 101.328145] entry_SYSCALL_64_fastpath+0x1f/0xbe
> [ 101.328149]
> [ 101.328154] Chain exists of:
> [ 101.328169] Possible unsafe locking scenario:
> [ 101.328174] CPU0 CPU1
> [ 101.328177] ---- ----
> [ 101.328180] lock(&ei->dio_sem);
> [ 101.328187] lock(&ei->log_mutex);
> [ 101.328194] lock(&ei->dio_sem);
> [ 101.328200] lock(&mm->mmap_sem);
> [ 101.328206]
> [ 101.328213] 2 locks held by mysqld/1253:
> [ 101.328217] #0: (sb_writers#10){.+.+}, at: [<ffffffffbd34a681>]
> aio_write+0x191/0x1a0
> [ 101.328231] #1: (&ei->dio_sem){++++}, at: [<ffffffffc02e8bdf>]
> btrfs_direct_IO+0x39f/0x400 [btrfs]
> [ 101.328277]
> [ 101.328285] CPU: 0 PID: 1253 Comm: mysqld Not tainted
> 4.14.0-0.rc3.git1.1.fc28.x86_64 #1
> [ 101.328290] Hardware name: Sony Corporation VPCSB2M9E/VAIO, BIOS
> R2087H4 06/15/2012
> [ 101.328294] Call Trace:
> [ 101.328304] dump_stack+0x8e/0xd6
> [ 101.328314] print_circular_bug+0x1f6/0x2e0
> [ 101.328322] ? copy_trace+0x90/0x90
> [ 101.328330] check_prev_add+0x351/0x700
> [ 101.328341] ? rcu_read_lock_sched_held+0x22/0x80
> [ 101.328348] ? __module_address+0x2c/0xf0
> [ 101.328361] __lock_acquire+0x1107/0x11d0
> [ 101.328368] ? __lock_acquire+0x1107/0x11d0
> [ 101.328376] ? copy_trace+0x90/0x90
> [ 101.328387] lock_acquire+0xa3/0x1f0
> [ 101.328395] ? get_user_pages_unlocked+0x5e/0x1b0
> [ 101.328406] down_read+0x48/0xb0
> [ 101.328412] ? get_user_pages_unlocked+0x5e/0x1b0
> [ 101.328419] get_user_pages_unlocked+0x5e/0x1b0
> [ 101.328429] get_user_pages_fast+0x7a/0xc0
> [ 101.328439] iov_iter_get_pages+0xc9/0x300
> [ 101.328450] do_blockdev_direct_IO+0x192b/0x2940
> [ 101.328459] ? update_stack_state+0x144/0x160
> [ 101.328501] ? btrfs_endio_direct_read+0xc0/0xc0 [btrfs]
> [ 101.328544] ? btrfs_page_exists_in_range+0x310/0x310 [btrfs]
> [ 101.328590] ? btrfs_page_exists_in_range+0x310/0x310 [btrfs]
> [ 101.328628] ? btrfs_endio_direct_read+0xc0/0xc0 [btrfs]
> [ 101.328642] __blockdev_direct_IO+0x2e/0x30
> [ 101.328649] ? __blockdev_direct_IO+0x2e/0x30
> [ 101.328687] ? btrfs_endio_direct_read+0xc0/0xc0 [btrfs]
> [ 101.328724] btrfs_direct_IO+0x171/0x400 [btrfs]
> [ 101.328760] ? btrfs_endio_direct_read+0xc0/0xc0 [btrfs]
> [ 101.328772] generic_file_direct_write+0xa3/0x160
> [ 101.328816] btrfs_file_write_iter+0x2fb/0x610 [btrfs]
> [ 101.328824] ? rcu_read_lock_sched_held+0x79/0x80
> [ 101.328833] aio_write+0x116/0x1a0
> [ 101.328843] ? sched_clock+0x9/0x10
> [ 101.328851] ? __might_fault+0x3e/0x90
> [ 101.328860] do_io_submit+0x42d/0x940
> [ 101.328865] ? do_io_submit+0x42d/0x940
> [ 101.328877] SyS_io_submit+0x10/0x20
> [ 101.328882] ? SyS_io_submit+0x10/0x20
> [ 101.328889] entry_SYSCALL_64_fastpath+0x1f/0xbe
> [ 101.328894] RIP: 0033:0x7f14cb326687
> [ 101.328899] RSP: 002b:00007f14aece44a8 EFLAGS: 00000206 ORIG_RAX:
> 00000000000000d1
> [ 101.328907] RAX: ffffffffffffffda RBX: 00007f14b4698000 RCX: 00007f14cb326687
> [ 101.328912] RDX: 00007f14aece44b0 RSI: 0000000000000001 RDI: 00007f14c2e26000
> [ 101.328916] RBP: 00007f14aece4830 R08: 00007f14aece44b0 R09: 0000000000000400
> [ 101.328920] R10: 0000000000000000 R11: 0000000000000206 R12: 00007f14b3d18630
> [ 101.328925] R13: 000055f10ab6ef48 R14: 00000012406e5969 R15: 0000000000000001
>
>
> kloczek
> --
> Tomasz Kłoczko | LinkedIn: http://lnkd.in/FXPWxH
>
>
> On 6 October 2017 at 16:34, David Sterba <dsterba@suse.com> wrote:
> > Hi,
> >
> > we have two more fixes for bugs introduced in 4.13. The sector_t problem with
> > 32bit architecture and !LBDAF config seems serious but the number of affected
> > deployments is hopefully low. The clashing status bits could lead to a
> > confusing in-memory state of the whole-filesystem operations if used with the
> > quota override sysfs knob.
> >
> > Please pull, thanks.
> >
> > The following changes since commit 8c6c592831a09a28428448e68fb08c6bbb8b9b8b:
> >
> > btrfs: log csums for all modified extents (2017-09-26 14:54:16 +0200)
> >
> > are available in the git repository at:
> >
> > git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux.git for-4.14-rc4
> >
> > for you to fetch changes up to 69ad59767d094752c23c0fc180a79532fde073d0:
> >
> > Btrfs: fix overlap of fs_info::flags values (2017-10-04 16:44:18 +0200)
> >
> > ----------------------------------------------------------------
> > Goffredo Baroncelli (1):
> > btrfs: avoid overflow when sector_t is 32 bit
> >
> > Tsutomu Itoh (1):
> > Btrfs: fix overlap of fs_info::flags values
> >
> > fs/btrfs/ctree.h | 2 +-
> > fs/btrfs/extent_io.c | 2 +-
> > 2 files changed, 2 insertions(+), 2 deletions(-)
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
prev parent reply other threads:[~2017-10-06 23:28 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-06 15:34 [GIT PULL] Btrfs fixes for 4.14-rc4 David Sterba
2017-10-06 22:25 ` Tomasz Kłoczko
2017-10-06 23:23 ` Liu Bo [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=20171006232339.GA19068@lim.localdomain \
--to=bo.li.liu@oracle.com \
--cc=clm@fb.com \
--cc=dsterba@suse.com \
--cc=kloczko.tomasz@gmail.com \
--cc=linux-btrfs@vger.kernel.org \
/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;
as well as URLs for NNTP newsgroup(s).