Re: [bug report] [PATCH v6] net: usb: Convert tasklet API to new bottom half workqueue mechanism

[qiang.zhang@linux.dev]

> 
> > 
> > Hi Jun,
> > 
> >  
> > 
> >  On Wed, Jun 18, 2025 at 01:39:23PM -0400, Jun Miao wrote:
> > 
> >  
> > 
> >  
> > 
> >  Migrate tasklet APIs to the new bottom half workqueue mechanism. It
> > 
> >  
> > 
> >  replaces all occurrences of tasklet usage with the appropriate workqueue
> > 
> >  
> > 
> >  APIs throughout the usbnet driver. This transition ensures compatibility
> > 
> >  
> > 
> >  with the latest design and enhances performance.
> > 
> >  
> > 
> >  
> > 
> >  After applying this patch, the smsc95xx driver fails after one down/up
> > 
> >  
> > 
> >  cycle.
> > 
> 
> Hello, Oleksij
> 
> Please try follow patch base on Jun Miao's patchs:
> 

Sorry, please ignore previous, try it:

diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index 9564478a79cc..6a3cca104af9 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -861,14 +861,14 @@ int usbnet_stop (struct net_device *net)
        /* deferred work (timer, softirq, task) must also stop */
        dev->flags = 0;
        timer_delete_sync(&dev->delay);
-       disable_work_sync(&dev->bh_work);
+       cancel_work_sync(&dev->bh_work);
        cancel_work_sync(&dev->kevent);
 
        /* We have cyclic dependencies. Those calls are needed
         * to break a cycle. We cannot fall into the gaps because
         * we have a flag
         */
-       disable_work_sync(&dev->bh_work);
+       cancel_work_sync(&dev->bh_work);
        timer_delete_sync(&dev->delay);
        cancel_work_sync(&dev->kevent);


> diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
> 
> index 9564478a79cc..554f1a1cf247 100644
> 
> --- a/drivers/net/usb/usbnet.c
> 
> +++ b/drivers/net/usb/usbnet.c
> 
> @@ -953,6 +953,7 @@ int usbnet_open (struct net_device *net)
> 
>  dev->pkt_cnt = 0;
> 
>  dev->pkt_err = 0;
> 
>  clear_bit(EVENT_RX_KILL, &dev->flags);
> 
> + enable_work(&dev->bh_work);
> 
>  
> 
>  // delay posting reads until we're fully open
> 
>  queue_work(system_bh_wq, &dev->bh_work);
> 
> Thanks
> 
> Zqiang
> 
> > 
> > Here is how I can reproduce the issue:
> > 
> >  
> > 
> >  nmcli device set enu1u1 managed no
> > 
> >  
> > 
> >  ip a a 10.10.10.1/24 dev enu1u1
> > 
> >  
> > 
> >  ping -c 4 10.10.10.3
> > 
> >  
> > 
> >  ip l s dev enu1u1 down
> > 
> >  
> > 
> >  ip l s dev enu1u1 up
> > 
> >  
> > 
> >  ping -c 4 10.10.10.3
> > 
> >  
> > 
> >  The second ping does not reach the host. Networking also fails on other
> > 
> >  
> > 
> >  interfaces.
> > 
> >  
> > 
> >  After some delay, the following trace appears:
> > 
> >  
> > 
> >  [ 846.838527] INFO: task kworker/u16:1:308 blocked for more than 120 seconds.
> > 
> >  
> > 
> >  [ 846.838596] Not tainted 6.16.0-rc3-00963-g4fcedea9cdf2-dirty #32
> > 
> >  
> > 
> >  [ 846.838666] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> > 
> >  
> > 
> >  [ 846.838697] task:kworker/u16:1 state:D stack:0 pid:308 tgid:308 ppid:2
> > 
> >  
> > 
> >  task_flags:0x4208060 flags:0x00000010
> > 
> >  
> > 
> >  [ 846.838776] Workqueue: events_unbound linkwatch_event
> > 
> >  
> > 
> >  [ 846.838851] Call trace:
> > 
> >  
> > 
> >  [ 846.838880] __switch_to+0x1d0/0x330 (T)
> > 
> >  
> > 
> >  [ 846.838933] __schedule+0xa88/0x2a90
> > 
> >  
> > 
> >  [ 846.838980] schedule+0x114/0x428
> > 
> >  
> > 
> >  [ 846.839010] schedule_preempt_disabled+0x80/0x118
> > 
> >  
> > 
> >  [ 846.839036] __mutex_lock+0x764/0xba8
> > 
> >  
> > 
> >  [ 846.839060] mutex_lock_nested+0x28/0x38
> > 
> >  
> > 
> >  [ 846.839084] rtnl_lock+0x20/0x30
> > 
> >  
> > 
> >  [ 846.839115] linkwatch_event+0x18/0x70
> > 
> >  
> > 
> >  [ 846.839141] process_one_work+0x760/0x17b0
> > 
> >  
> > 
> >  [ 846.839175] worker_thread+0x768/0xce8
> > 
> >  
> > 
> >  [ 846.839208] kthread+0x3bc/0x690
> > 
> >  
> > 
> >  [ 846.839237] ret_from_fork+0x10/0x20
> > 
> >  
> > 
> >  [ 846.839359] INFO: task kworker/u16:1:308 is blocked on a mutex likely
> > 
> >  
> > 
> >  owned by task ip:899.
> > 
> >  
> > 
> >  [ 846.839381] task:ip state:D stack:0 pid:899
> > 
> >  
> > 
> >  tgid:899 ppid:1 task_flags:0x400100 flags:0x00000019
> > 
> >  
> > 
> >  [ 846.839419] Call trace:
> > 
> >  
> > 
> >  [ 846.839432] __switch_to+0x1d0/0x330 (T)
> > 
> >  
> > 
> >  [ 846.839466] __schedule+0xa88/0x2a90
> > 
> >  
> > 
> >  [ 846.839495] schedule+0x114/0x428
> > 
> >  
> > 
> >  [ 846.839524] schedule_timeout+0xec/0x220
> > 
> >  
> > 
> >  [ 846.839551] wait_skb_queue_empty+0xa0/0x168
> > 
> >  
> > 
> >  [ 846.839581] usbnet_terminate_urbs+0x150/0x2c8
> > 
> >  
> > 
> >  [ 846.839609] usbnet_stop+0x41c/0x608
> > 
> >  
> > 
> >  [ 846.839636] __dev_close_many+0x1fc/0x4b8
> > 
> >  
> > 
> >  [ 846.839668] __dev_change_flags+0x33c/0x500
> > 
> >  
> > 
> >  [ 846.839694] netif_change_flags+0x7c/0x158
> > 
> >  
> > 
> >  [ 846.839718] do_setlink.isra.0+0x2040/0x2eb8
> > 
> >  
> > 
> >  [ 846.839745] rtnl_newlink+0xd88/0x16c8
> > 
> >  
> > 
> >  [ 846.839770] rtnetlink_rcv_msg+0x654/0x8c8
> > 
> >  
> > 
> >  [ 846.839795] netlink_rcv_skb+0x19c/0x350
> > 
> >  
> > 
> >  [ 846.839823] rtnetlink_rcv+0x1c/0x30
> > 
> >  
> > 
> >  [ 846.839848] netlink_unicast+0x3c4/0x668
> > 
> >  
> > 
> >  [ 846.839873] netlink_sendmsg+0x620/0xa10
> > 
> >  
> > 
> >  [ 846.839899] ____sys_sendmsg+0x2f8/0x788
> > 
> >  
> > 
> >  [ 846.839924] ___sys_sendmsg+0xf0/0x178
> > 
> >  
> > 
> >  [ 846.839950] __sys_sendmsg+0x104/0x198
> > 
> >  
> > 
> >  [ 846.839975] __arm64_sys_sendmsg+0x74/0xa8
> > 
> >  
> > 
> >  [ 846.840000] el0_svc_common.constprop.0+0xe4/0x338
> > 
> >  
> > 
> >  [ 846.840033] do_el0_svc+0x44/0x60
> > 
> >  
> > 
> >  [ 846.840061] el0_svc+0x3c/0xb0
> > 
> >  
> > 
> >  [ 846.840089] el0t_64_sync_handler+0x104/0x130
> > 
> >  
> > 
> >  [ 846.840117] el0t_64_sync+0x154/0x158
> > 
> >  
> > 
> >  [ 846.840164]
> > 
> >  
> > 
> >  [ 846.840164] Showing all locks held in the system:
> > 
> >  
> > 
> >  [ 846.840199] 1 lock held by khungtaskd/41:
> > 
> >  
> > 
> >  [ 846.840216] #0: ffffffc08424ede0 (rcu_read_lock){....}-{1:3}, at: debug_show_all_locks+0x14/0x1b0
> > 
> >  
> > 
> >  [ 846.840309] 3 locks held by kworker/u16:2/47:
> > 
> >  
> > 
> >  [ 846.840325] #0: ffffff800926a148 ((wq_completion)ipv6_addrconf){+.+.}-{0:0}, at: process_one_work+0x698/0x17b0
> > 
> >  
> > 
> >  [ 846.840406] #1: ffffffc0860f7c00 ((work_completion)(&(&net->ipv6.addr_chk_work)->work)){+.+.}-{0:0}, at: process_one_work+0x6bc/0x17b0
> > 
> >  
> > 
> >  [ 846.840484] #2: ffffffc084924408 (rtnl_mutex){+.+.}-{4:4}, at: rtnl_lock+0x20/0x30
> > 
> >  
> > 
> >  [ 846.840568] 2 locks held by pr/ttymxc1/60:
> > 
> >  
> > 
> >  [ 846.840595] 5 locks held by sugov:0/84:
> > 
> >  
> > 
> >  [ 846.840618] 2 locks held by systemd-journal/124:
> > 
> >  
> > 
> >  [ 846.840639] 3 locks held by kworker/u16:1/308:
> > 
> >  
> > 
> >  [ 846.840655] #0: ffffff8005d00148 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x698/0x17b0
> > 
> >  
> > 
> >  [ 846.840733] #1: ffffffc087307c00 ((linkwatch_work).work){+.+.}-{0:0}, at: process_one_work+0x6bc/0x17b0
> > 
> >  
> > 
> >  [ 846.840810] #2: ffffffc084924408 (rtnl_mutex){+.+.}-{4:4}, at: rtnl_lock+0x20/0x30
> > 
> >  
> > 
> >  [ 846.840894] 1 lock held by ip/899:
> > 
> >  
> > 
> >  [ 846.840910] #0: ffffffc084924408 (rtnl_mutex){+.+.}-{4:4}, at: rtnl_newlink+0x5e8/0x16c8
> > 
> >  
> > 
> >  [ 846.840982] 2 locks held by sshd/901:
> > 
> >  
> > 
> >  [ 846.840998] #0: ffffff800aee06b0 (nlk_cb_mutex-ROUTE){+.+.}-{4:4}, at: __netlink_dump_start+0x100/0x800
> > 
> >  
> > 
> >  [ 846.841073] #1: ffffffc084924408 (rtnl_mutex){+.+.}-{4:4}, at: rtnl_dumpit+0x128/0x1a8
> > 
> >  
> > 
> >  [ 846.841149] 2 locks held by sshd/903:
> > 
> >  
> > 
> >  [ 846.841165] #0: ffffff800b1c76b0 (nlk_cb_mutex-ROUTE){+.+.}-{4:4}, at: __netlink_dump_start+0x100/0x800
> > 
> >  
> > 
> >  [ 846.841237] #1: ffffffc084924408 (rtnl_mutex){+.+.}-{4:4}, at: rtnl_dumpit+0x128/0x1a8
> > 
> >  
> > 
> >  Reverting this patch recovers smsc95xx functionality.
> > 
> >  
> > 
> >  Best Regards,
> > 
> >  
> > 
> >  Oleksij
> >
>