From: Thomas Gleixner <tglx@linutronix.de>
To: Jens Axboe <axboe@kernel.dk>
Cc: linux-fsdevel@vger.kernel.org,
Alexander Viro <viro@zeniv.linux.org.uk>,
Pavel Begunkov <asml.silence@gmail.com>,
io-uring@vger.kernel.org,
Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Subject: [BUG] io-uring triggered lockdep splat
Date: Tue, 10 Aug 2021 09:57:26 +0200 [thread overview]
Message-ID: <87r1f1speh.ffs@tglx> (raw)
Jens,
running 'rsrc_tags' from the liburing tests on v5.14-rc5 triggers the
following lockdep splat:
[ 265.866713] ======================================================
[ 265.867585] WARNING: possible circular locking dependency detected
[ 265.868450] 5.14.0-rc5 #69 Tainted: G E
[ 265.869174] ------------------------------------------------------
[ 265.870050] kworker/3:1/86 is trying to acquire lock:
[ 265.870759] ffff88812100f0a8 (&ctx->uring_lock){+.+.}-{3:3}, at: io_rsrc_put_work+0x142/0x1b0
[ 265.871957]
but task is already holding lock:
[ 265.872777] ffffc900004a3e70 ((work_completion)(&(&ctx->rsrc_put_work)->work)){+.+.}-{0:0}, at: process_one_work+0x218/0x590
[ 265.874334]
which lock already depends on the new lock.
[ 265.875474]
the existing dependency chain (in reverse order) is:
[ 265.876512]
-> #1 ((work_completion)(&(&ctx->rsrc_put_work)->work)){+.+.}-{0:0}:
[ 265.877750] __flush_work+0x372/0x4f0
[ 265.878343] io_rsrc_ref_quiesce.part.0.constprop.0+0x35/0xb0
[ 265.879227] __do_sys_io_uring_register+0x652/0x1080
[ 265.880009] do_syscall_64+0x3b/0x90
[ 265.880598] entry_SYSCALL_64_after_hwframe+0x44/0xae
[ 265.881383]
-> #0 (&ctx->uring_lock){+.+.}-{3:3}:
[ 265.882257] __lock_acquire+0x1130/0x1df0
[ 265.882903] lock_acquire+0xc8/0x2d0
[ 265.883485] __mutex_lock+0x88/0x780
[ 265.884067] io_rsrc_put_work+0x142/0x1b0
[ 265.884713] process_one_work+0x2a2/0x590
[ 265.885357] worker_thread+0x55/0x3c0
[ 265.885958] kthread+0x143/0x160
[ 265.886493] ret_from_fork+0x22/0x30
[ 265.887079]
other info that might help us debug this:
[ 265.888206] Possible unsafe locking scenario:
[ 265.889043] CPU0 CPU1
[ 265.889687] ---- ----
[ 265.890328] lock((work_completion)(&(&ctx->rsrc_put_work)->work));
[ 265.891211] lock(&ctx->uring_lock);
[ 265.892074] lock((work_completion)(&(&ctx->rsrc_put_work)->work));
[ 265.893310] lock(&ctx->uring_lock);
[ 265.893833]
*** DEADLOCK ***
[ 265.894660] 2 locks held by kworker/3:1/86:
[ 265.895252] #0: ffff888100059738 ((wq_completion)events){+.+.}-{0:0}, at: process_one_work+0x218/0x590
[ 265.896561] #1: ffffc900004a3e70 ((work_completion)(&(&ctx->rsrc_put_work)->work)){+.+.}-{0:0}, at: process_one_work+0x218/0x590
[ 265.898178]
stack backtrace:
[ 265.898789] CPU: 3 PID: 86 Comm: kworker/3:1 Kdump: loaded Tainted: G E 5.14.0-rc5 #69
[ 265.900072] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014
[ 265.901195] Workqueue: events io_rsrc_put_work
[ 265.901825] Call Trace:
[ 265.902173] dump_stack_lvl+0x57/0x72
[ 265.902698] check_noncircular+0xf2/0x110
[ 265.903270] ? __lock_acquire+0x380/0x1df0
[ 265.903889] __lock_acquire+0x1130/0x1df0
[ 265.904462] lock_acquire+0xc8/0x2d0
[ 265.904967] ? io_rsrc_put_work+0x142/0x1b0
[ 265.905596] ? lock_is_held_type+0xa5/0x120
[ 265.906193] __mutex_lock+0x88/0x780
[ 265.906700] ? io_rsrc_put_work+0x142/0x1b0
[ 265.907286] ? io_rsrc_put_work+0x142/0x1b0
[ 265.907877] ? lock_acquire+0xc8/0x2d0
[ 265.908408] io_rsrc_put_work+0x142/0x1b0
[ 265.908976] process_one_work+0x2a2/0x590
[ 265.909544] worker_thread+0x55/0x3c0
[ 265.910061] ? process_one_work+0x590/0x590
[ 265.910655] kthread+0x143/0x160
[ 265.911114] ? set_kthread_struct+0x40/0x40
[ 265.911704] ret_from_fork+0x22/0x30
Thanks,
tglx
next reply other threads:[~2021-08-10 7:57 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-10 7:57 Thomas Gleixner [this message]
2021-08-10 9:58 ` [BUG] io-uring triggered lockdep splat Pavel Begunkov
2021-08-10 16:51 ` Thomas Gleixner
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=87r1f1speh.ffs@tglx \
--to=tglx@linutronix.de \
--cc=asml.silence@gmail.com \
--cc=axboe@kernel.dk \
--cc=bigeasy@linutronix.de \
--cc=io-uring@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=viro@zeniv.linux.org.uk \
/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).