Re: BUG: selftest/net/tun: Hang in unregister_netdevice

From: Kuniyuki Iwashima
Date: Wed Mar 15 2023 - 16:57:18 EST


From: Mirsad Goran Todorovac <mirsad.todorovac@xxxxxxxxxxxx>
Date: Tue, 14 Mar 2023 21:10:55 +0100
> On 14. 03. 2023. 17:02, Eric Dumazet wrote:
> > On Tue, Mar 14, 2023 at 9:01 AM Mirsad Todorovac
> > <mirsad.todorovac@xxxxxxxxxxxx> wrote:
> >
> >> After a while, kernel message start looping:
> >>
> >> kernel:unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> >>
> >> Message from syslogd@pc-mtodorov at Mar 14 16:57:15 ...
> >> kernel:unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> >>
> >> Message from syslogd@pc-mtodorov at Mar 14 16:57:24 ...
> >> kernel:unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> >>
> >> Message from syslogd@pc-mtodorov at Mar 14 16:57:26 ...
> >> kernel:unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> >>
> >> This hangs processes until very late stage of shutdown.
> >>
> >> I can confirm that CONFIG_DEBUG_{KOBJECT,KOBJECT_RELEASE}=y were the only changes
> >> to .config in between builds.

KOBJECT_RELEASE delays freeing kobject with delayed_work.

In the failing tests, e.g. reattach_delete_close, we create
a tun dev and do

1. detach (tun_detach())
2. re-attach (tun_attach())
3. close

. We release kobjects (tun->dev->(_tx|_rx).kobj) in 1. and
recycle the same memory region in 2.

But the kobjects are not actually released due to the delay,
thus netdev_queue_add_kobject() and rx_queue_add_kobject()
fail.

---8<---
# RUN tun.reattach_delete_close ...
[ 179.087589] kobject: 'tx-1' (00000000ee182e2f): kobject_release, parent 000000000643eb5f (delayed 3000)
[ 179.088105] kobject: 'rx-1' (000000001c44852d): kobject_release, parent 000000000643eb5f (delayed 1000)
[ 179.089097] kobject (00000000ee182e2f): tried to init an initialized object, something is seriously wrong.
---8<---

However, we don't assume the delay and also the failure in
tun_set_real_num_queues().

In this case, we have to re-initialise the queues without
touching kobjects.

Eric,
Are you working on this?
If not, let me try fixing this :)

Thanks,
Kuniyuki


> >>
> >> Best regards,
> >> Mirsad
> >>
> >
> > Try adding in your config
> >
> > CONFIG_NET_DEV_REFCNT_TRACKER=y
> > CONFIG_NET_NS_REFCNT_TRACKER=y
> >
> > Thanks.
>
> Not at all.
>
> According to the info here: https://cateee.net/lkddb/web-lkddb/NET_DEV_REFCNT_TRACKER.html
> no kerenel param was needed.
>
> I have got the same hang, and additional debug information appears to be this
> (in /var/log/messages):
>
> Mar 14 20:58:20 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:58:20 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:20 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:58:20 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:58:20 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:58:20 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:20 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:20 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:20 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:20 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:20 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:20 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:58:20 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:58:20 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:58:20 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:20 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:20 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:20 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:20 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:20 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:58:20 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:20 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:58:20 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:58:20 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:58:20 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:20 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:20 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:20 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:20 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:20 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:20 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:58:20 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:58:20 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:58:20 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:20 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:20 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:20 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:20 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:30 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:58:30 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:30 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:58:30 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:58:30 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:58:30 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:30 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:30 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:30 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:30 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:30 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:30 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:58:30 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:58:30 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:58:30 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:30 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:30 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:30 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:30 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:30 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:58:30 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:30 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:58:30 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:58:30 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:58:30 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:30 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:30 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:30 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:30 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:30 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:30 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:58:30 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:58:30 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:58:30 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:30 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:30 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:30 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:30 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:40 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:58:40 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:40 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:58:40 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:58:40 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:58:40 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:40 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:40 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:40 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:40 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:40 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:40 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:58:40 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:58:40 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:58:40 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:40 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:40 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:40 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:40 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:40 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:58:40 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:40 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:58:40 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:58:40 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:58:40 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:40 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:40 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:40 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:40 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:40 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:40 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:58:40 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:58:40 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:58:40 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:40 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:40 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:40 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:40 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:50 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:58:50 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:50 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:58:50 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:58:50 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:58:50 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:50 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:50 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:50 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:50 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:50 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:50 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:58:50 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:58:50 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:58:50 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:50 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:50 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:50 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:50 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:50 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:58:50 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:50 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:58:50 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:58:50 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:58:50 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:50 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:50 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:50 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:50 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:50 pc-mtodorov kernel: leaked reference.
> Mar 14 20:58:50 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:58:50 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:58:50 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:58:50 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:58:50 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:58:50 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:58:50 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:58:50 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:58:57 pc-mtodorov kernel: kmemleak: 1 new suspected memory leaks (see /sys/kernel/debug/kmemleak)
> Mar 14 20:59:00 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:59:00 pc-mtodorov kernel: leaked reference.
> Mar 14 20:59:00 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:59:00 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:59:00 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:59:00 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:59:00 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:59:00 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:59:00 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:59:00 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:59:00 pc-mtodorov kernel: leaked reference.
> Mar 14 20:59:00 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:59:00 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:59:00 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:59:00 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:59:00 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:59:00 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:59:00 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:59:00 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:59:01 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:59:01 pc-mtodorov kernel: leaked reference.
> Mar 14 20:59:01 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:59:01 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:59:01 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:59:01 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:59:01 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:59:01 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:59:01 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:59:01 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:59:01 pc-mtodorov kernel: leaked reference.
> Mar 14 20:59:01 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:59:01 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:59:01 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:59:01 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:59:01 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:59:01 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:59:01 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:59:01 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:59:10 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:59:10 pc-mtodorov kernel: leaked reference.
> Mar 14 20:59:10 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:59:10 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:59:10 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:59:10 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:59:10 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:59:10 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:59:10 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:59:10 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:59:10 pc-mtodorov kernel: leaked reference.
> Mar 14 20:59:10 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:59:10 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:59:10 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:59:10 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:59:10 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:59:10 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:59:10 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:59:10 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:59:11 pc-mtodorov kernel: unregister_netdevice: waiting for tap0 to become free. Usage count = 3
> Mar 14 20:59:11 pc-mtodorov kernel: leaked reference.
> Mar 14 20:59:11 pc-mtodorov kernel: net_rx_queue_update_kobjects+0x75/0x1d0
> Mar 14 20:59:11 pc-mtodorov kernel: netif_set_real_num_rx_queues+0x5b/0xb0
> Mar 14 20:59:11 pc-mtodorov kernel: tun_attach+0x1ec/0x5a0
> Mar 14 20:59:11 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:59:11 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:59:11 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:59:11 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:59:11 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> Mar 14 20:59:11 pc-mtodorov kernel: leaked reference.
> Mar 14 20:59:11 pc-mtodorov kernel: netdev_queue_update_kobjects+0x86/0x190
> Mar 14 20:59:11 pc-mtodorov kernel: netif_set_real_num_tx_queues+0x86/0x250
> Mar 14 20:59:11 pc-mtodorov kernel: tun_attach+0x1d7/0x5a0
> Mar 14 20:59:11 pc-mtodorov kernel: __tun_chr_ioctl+0xa58/0x17d0
> Mar 14 20:59:11 pc-mtodorov kernel: tun_chr_ioctl+0x17/0x20
> Mar 14 20:59:11 pc-mtodorov kernel: __x64_sys_ioctl+0x97/0xd0
> Mar 14 20:59:11 pc-mtodorov kernel: do_syscall_64+0x5c/0x90
> Mar 14 20:59:11 pc-mtodorov kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc
> [root@pc-mtodorov marvin]#
>
> I see those "leaked reference" lines are being printed here:
> https://elixir.bootlin.com/linux/v6.3-rc2/source/lib/ref_tracker.c#L55
>
> However, it is beyond the scope of my knowledge to track the actual leak.
>
> Hope this helps.
>
> Best regards,
> Mirsad
>
> --
> Mirsad Goran Todorovac
> Sistem inženjer
> Grafički fakultet | Akademija likovnih umjetnosti
> Sveučilište u Zagrebu
>
> System engineer
> Faculty of Graphic Arts | Academy of Fine Arts
> University of Zagreb, Republic of Croatia
> The European Union