From: Bart Van Assche <bvanassche@acm.org>
To: Bob Pearson <rpearsonhpe@gmail.com>
Cc: "linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>
Subject: rdma-for-next, rdma_rxe: inconsistent lock state
Date: Tue, 31 May 2022 13:46:41 -0700 [thread overview]
Message-ID: <24ed4fac-0519-be62-6817-b4a73050e805@acm.org> (raw)
Hi Bob,
With the rdma-for-next branch (commit 9c477178a0a1 ("RDMA/rtrs-clt: Fix one
kernel-doc comment")) I see the following:
================================
WARNING: inconsistent lock state
5.18.0-dbg #4 Not tainted
--------------------------------
inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage.
ksoftirqd/2/25 [HC0[0]:SC1[1]:HE0:SE0] takes:
ffff888116f0d350 (&xa->xa_lock#12){+.?.}-{2:2}, at: rxe_pool_get_index+0x73/0x170 [rdma_rxe]
{SOFTIRQ-ON-W} state was registered at:
__lock_acquire+0x45b/0xce0
lock_acquire+0x18a/0x450
_raw_spin_lock+0x34/0x50
__rxe_add_to_pool+0xcc/0x140 [rdma_rxe]
rxe_alloc_pd+0x2d/0x40 [rdma_rxe]
__ib_alloc_pd+0xa3/0x270 [ib_core]
ib_mad_port_open+0x44a/0x790 [ib_core]
ib_mad_init_device+0x8e/0x110 [ib_core]
add_client_context+0x26a/0x330 [ib_core]
enable_device_and_get+0x169/0x2b0 [ib_core]
ib_register_device+0x26f/0x330 [ib_core]
rxe_register_device+0x1b4/0x1d0 [rdma_rxe]
rxe_add+0x8c/0xc0 [rdma_rxe]
rxe_net_add+0x5b/0x90 [rdma_rxe]
rxe_newlink+0x71/0x80 [rdma_rxe]
nldev_newlink+0x21e/0x370 [ib_core]
rdma_nl_rcv_msg+0x200/0x410 [ib_core]
rdma_nl_rcv+0x140/0x220 [ib_core]
netlink_unicast+0x307/0x460
netlink_sendmsg+0x422/0x750
__sys_sendto+0x1c2/0x250
__x64_sys_sendto+0x7f/0x90
do_syscall_64+0x35/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
irq event stamp: 71543
hardirqs last enabled at (71542): [<ffffffff810cdc28>] __local_bh_enable_ip+0x88/0xf0
hardirqs last disabled at (71543): [<ffffffff81e9d67d>] _raw_spin_lock_irqsave+0x5d/0x60
softirqs last enabled at (71532): [<ffffffff82200467>] __do_softirq+0x467/0x6e1
softirqs last disabled at (71537): [<ffffffff810cda47>] run_ksoftirqd+0x37/0x60
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(&xa->xa_lock#12);
<Interrupt>
lock(&xa->xa_lock#12);
*** DEADLOCK ***
no locks held by ksoftirqd/2/25.
stack backtrace:
CPU: 2 PID: 25 Comm: ksoftirqd/2 Not tainted 5.18.0-dbg #4
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.15.0-0-g2dd4b9b-rebuilt.opensuse.org 04/01/2014
Call Trace:
<TASK>
show_stack+0x52/0x58
dump_stack_lvl+0x5b/0x82
dump_stack+0x10/0x12
print_usage_bug.part.0+0x29c/0x2ab
mark_lock_irq.cold+0x54/0xbf
mark_lock.part.0+0x3f5/0xa70
mark_usage+0x74/0x1a0
__lock_acquire+0x45b/0xce0
lock_acquire+0x18a/0x450
_raw_spin_lock_irqsave+0x43/0x60
rxe_pool_get_index+0x73/0x170 [rdma_rxe]
rxe_get_av+0xcc/0x140 [rdma_rxe]
rxe_requester+0x34c/0xe60 [rdma_rxe]
rxe_do_task+0xcc/0x140 [rdma_rxe]
tasklet_action_common.constprop.0+0x168/0x1b0
tasklet_action+0x42/0x60
__do_softirq+0x1d8/0x6e1
run_ksoftirqd+0x37/0x60
smpboot_thread_fn+0x302/0x410
kthread+0x183/0x1c0
ret_from_fork+0x1f/0x30
</TASK>
Is this perhaps the same issue as what I reported on May 6
(https://lore.kernel.org/all/cf8b9980-3965-a4f6-07e0-d4b25755b0db@acm.org/)?
Thanks,
Bart.
next reply other threads:[~2022-05-31 20:46 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-31 20:46 Bart Van Assche [this message]
2022-05-31 20:55 ` rdma-for-next, rdma_rxe: inconsistent lock state Pearson, Robert B
2022-05-31 22:24 ` Yanjun Zhu
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=24ed4fac-0519-be62-6817-b4a73050e805@acm.org \
--to=bvanassche@acm.org \
--cc=linux-rdma@vger.kernel.org \
--cc=rpearsonhpe@gmail.com \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.