From: Moksh Panicker <mokshpanicker.7@gmail.com>
To: kuba@kernel.org
Cc: andrew+netdev@lunn.ch, davem@davemloft.net, edumazet@google.com,
pabeni@redhat.com, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org, skhan@linuxfoundation.org,
xujiakai24@mails.ucas.ac.cn,
Moksh Panicker <mokshpanicker.7@gmail.com>,
syzbot+1cf303af03cf30b1275a@syzkaller.appspot.com
Subject: [PATCH v2] netdevsim: Fix deadlock in del_device_store() and nsim_bus_exit()
Date: Tue, 16 Jun 2026 22:39:04 +0000 [thread overview]
Message-ID: <20260616223904.42509-1-mokshpanicker.7@gmail.com> (raw)
In-Reply-To: <20260509092837.3432281-1-xujiakai24@mails.ucas.ac.cn>
del_device_store() holds nsim_bus_dev_list_lock while calling
nsim_bus_dev_del(), which calls device_unregister() which internally
acquires the device lock. Similarly, nsim_bus_exit() holds the same
lock while calling nsim_bus_dev_del(). If another thread already holds
the device lock and tries to acquire nsim_bus_dev_list_lock, a deadlock
occurs:
INFO: task hung in nsim_bus_dev_del
Fix this by releasing nsim_bus_dev_list_lock before calling
nsim_bus_dev_del() in both locations, after the devices have already
been removed from the list with list_del().
A similar issue exists in new_device_store() which can be addressed
separately.
Reported-by: syzbot+1cf303af03cf30b1275a@syzkaller.appspot.com
Closes: https://syzkaller.appspot.com/bug?extid=1cf303af03cf30b1275a
Signed-off-by: Moksh Panicker <mokshpanicker.7@gmail.com>
---
drivers/net/netdevsim/bus.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/drivers/net/netdevsim/bus.c b/drivers/net/netdevsim/bus.c
index 41483e371..0f02ff8ad 100644
--- a/drivers/net/netdevsim/bus.c
+++ b/drivers/net/netdevsim/bus.c
@@ -241,11 +241,12 @@ del_device_store(const struct bus_type *bus, const char *buf, size_t count)
if (nsim_bus_dev->dev.id != id)
continue;
list_del(&nsim_bus_dev->list);
- nsim_bus_dev_del(nsim_bus_dev);
err = 0;
break;
}
mutex_unlock(&nsim_bus_dev_list_lock);
+ if (!err)
+ nsim_bus_dev_del(nsim_bus_dev);
return !err ? count : err;
}
static BUS_ATTR_WO(del_device);
@@ -527,11 +528,11 @@ void nsim_bus_exit(void)
complete(&nsim_bus_devs_released);
mutex_lock(&nsim_bus_dev_list_lock);
- list_for_each_entry_safe(nsim_bus_dev, tmp, &nsim_bus_dev_list, list) {
+ list_for_each_entry_safe(nsim_bus_dev, tmp, &nsim_bus_dev_list, list)
list_del(&nsim_bus_dev->list);
- nsim_bus_dev_del(nsim_bus_dev);
- }
mutex_unlock(&nsim_bus_dev_list_lock);
+ list_for_each_entry_safe(nsim_bus_dev, tmp, &nsim_bus_dev_list, list)
+ nsim_bus_dev_del(nsim_bus_dev);
wait_for_completion(&nsim_bus_devs_released);
--
2.34.1
prev parent reply other threads:[~2026-06-16 22:39 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-09 9:28 [PATCH] netdevsim: Fix task hung by releasing bus lock before device ops Jiakai Xu
2026-05-09 9:33 ` Jiakai Xu
2026-05-10 3:03 ` kernel test robot
2026-05-10 5:18 ` kernel test robot
2026-05-10 5:41 ` kernel test robot
2026-06-16 22:39 ` Moksh Panicker [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=20260616223904.42509-1-mokshpanicker.7@gmail.com \
--to=mokshpanicker.7@gmail.com \
--cc=andrew+netdev@lunn.ch \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=kuba@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=skhan@linuxfoundation.org \
--cc=syzbot+1cf303af03cf30b1275a@syzkaller.appspot.com \
--cc=xujiakai24@mails.ucas.ac.cn \
/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