Re: REGRESSION: [PATCH] usb: typec: ucsi: Fix race condition and ordering in port unregistration

["Borah, Chaitanya Kumar" <chaitanya.kumar.borah@intel.com>]

On 7/16/2026 2:45 PM, Andrei Kuchynski wrote:
> On Thu, Jul 16, 2026 at 10:28 AM Borah, Chaitanya Kumar
> <chaitanya.kumar.borah@intel.com> wrote:
>>
>> Hello Andrei,
>>
>> On 7/7/2026 7:47 PM, Andrei Kuchynski wrote:
>>> A synchronization issue exists during port unregistration where pending
>>> partner work items can race against workqueue destruction, leading to
>>> use-after-free conditions:
>>>
>>>     cros_ec_ucsi cros_ec_ucsi.3.auto: error -ETIMEDOUT: PPM init failed
>>>     BUG: kernel NULL pointer dereference, address: 0000000000000000
>>>     RIP: 0010:__queue_work+0x83/0x4a0
>>>     Call Trace:
>>>       <IRQ>
>>>       __cfi_delayed_work_timer_fn+0x10/0x10
>>>       run_timer_softirq+0x3b6/0xbd0
>>>       sched_clock_cpu+0xc/0x110
>>>       irq_exit_rcu+0x18d/0x330
>>>       fred_sysvec_apic_timer_interrupt+0x5e/0x80
>>>
>>> Fix this by ensuring strict ordering and proper serialization during
>>> teardown:
>>>
>>> 1. Move ucsi_unregister_partner() to the beginning of the teardown
>>> sequence and protect it under the connector mutex lock.
>>> 2. Ensure all pending partner tasks are explicitly flushed and finished
>>> before the workqueue is destroyed.
>>> 3. Switch from mod_delayed_work() to a cancel_delayed_work() and
>>> queue_delayed_work() sequence. This guarantees that items currently marked
>>> as pending won't be scheduled an additional time, preventing a double
>>> release of resources which leads to the following crash:
>>>
>>>     Oops: general protection fault, probably for non-canonical address
>>>       0xdead000000000122: 0000 [#1] SMP NOPTI
>>>     Workqueue: cros_ec_ucsi.3.auto-con2 ucsi_poll_worker
>>>     RIP: 0010:ucsi_poll_worker+0x65/0x1e0
>>>     Call Trace:
>>>     <TASK>
>>>       process_scheduled_works+0x218/0x6d0
>>>       worker_thread+0x188/0x3f0
>>>       __cfi_worker_thread+0x10/0x10
>>>       kthread+0x226/0x2a0
>>>
>>> To ensure these rules are applied identically across both the normal
>>> teardown and the ucsi_init() error paths, consolidate the cleanup logic
>>> into a new helper, ucsi_unregister_port().
>>
>>
>> This seems to be causing regression in our linux-next CI. [1]
>>
>> <6>[   18.008557] ------------[ cut here ]------------
>> <4>[   18.013166] DEBUG_LOCKS_WARN_ON(1)
>> <4>[   18.013167] WARNING: kernel/locking/lockdep.c:238 at
>> __lock_acquire+0xa59/0x2750, CPU#1: kworker/u89:1/172
>> ...
>> <4>[   18.159950] Hardware name: Intel Corporation Meteor Lake Client
>> Platform/MTL-P DDR5 SODIMM SBS RVP, BIOS
>> MTLPFWI1.R00.3471.D89.2401091901 01/09/2024
>> <4>[   18.173125] Workqueue: events_dfl_long ucsi_init_work [typec_ucsi]
>> <4>[   18.179257] RIP: 0010:__lock_acquire+0xa60/0x2750
>> <4>[   18.183927] Code: 44 8b 5d a8 85 c0 0f 84 98 fd ff ff 44 8b 35 07
>> 0b 72 02 45 85 f6 0f 85 88 fd ff ff 48 8d 3d 27 25 73 02 48 c7 c6 1b 51
>> 27 83 <67> 48 0f b9 3a 31 c0 44 8b 5d a8 44 8b 4d b0 e9 c6 f7 ff ff 31 d2
>> ...
>> <4>[   18.263886] PKRU: 55555554
>> <4>[   18.266583] Call Trace:
>> <4>[   18.269014]  <TASK>
>> <4>[   18.271110]  ? mark_held_locks+0x46/0x90
>> <4>[   18.275007]  lock_acquire+0xd9/0x2e0
>> <4>[   18.278562]  ? ucsi_unregister_port+0x36/0x140 [typec_ucsi]
>> <4>[   18.284096]  __mutex_lock+0xb2/0x1020
>> <4>[   18.287735]  ? ucsi_unregister_port+0x36/0x140 [typec_ucsi]
>> <4>[   18.293265]  ? ucsi_unregister_port+0x36/0x140 [typec_ucsi]
>> <4>[   18.298797]  ? synchronize_rcu_expedited+0x2b2/0x300
>> <4>[   18.303724]  ? __pfx_autoremove_wake_function+0x10/0x10
>> <4>[   18.308912]  mutex_lock_nested+0x1b/0x30
>> <4>[   18.312806]  ? mutex_lock_nested+0x1b/0x30
>> <4>[   18.316878]  ucsi_unregister_port+0x36/0x140 [typec_ucsi]
>> <4>[   18.322238]  ucsi_init_work+0x66d/0xc10 [typec_ucsi]
>> <4>[   18.327167]  process_one_work+0x2ad/0x8b0
>> <4>[   18.331158]  worker_thread+0x200/0x3f0
>> <4>[   18.334886]  ? __pfx_worker_thread+0x10/0x10
>> <4>[   18.339128]  kthread+0x10d/0x150
>> <4>[   18.342339]  ? __pfx_kthread+0x10/0x10
>> <4>[   18.346067]  ret_from_fork+0x3bd/0x470
>> <4>[   18.349795]  ? __pfx_kthread+0x10/0x10
>> <4>[   18.353522]  ret_from_fork_asm+0x1a/0x30
>> <4>[   18.357420]  </TASK>
>>
>> Detailed log can be seen found in [2]
>>
>> We confirmed that reverting the patch solves the issue.
>>
>> Could you please check why the patch causes this regression and provide
>> a fix if necessary?
>>
>> Regards
>> Chaitanya
>>
>> [1] https://intel-gfx-ci.01.org/tree/linux-next/combined-alt.html?
>> [2]
>> https://intel-gfx-ci.01.org/tree/linux-next/next-20260714/bat-mtlp-9/boot0.txt
>>
>> -- Bisect Logs --
>>
>> git bisect start
>> # status: waiting for both good and bad commits
>> # good: [5f1b513690edf51727e6928b97705b6437f9a98e] Merge remote-tracking
>> branch 'spi/for-7.3' into spi-next
>> git bisect good 5f1b513690edf51727e6928b97705b6437f9a98e
>> # status: waiting for bad commit, 1 good commit known
>> # bad: [49362394dad7df66c274c867a271394c10ca2bb8] Add linux-next
>> specific files for 20260713
>> git bisect bad 49362394dad7df66c274c867a271394c10ca2bb8
>> # bad: [dffddeb046466ed57038295bde28b42b7506a12b] Merge branch
>> 'nand/next' of https://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux.git
>> git bisect bad dffddeb046466ed57038295bde28b42b7506a12b
>> # bad: [d154d0766bf3fe830e1bf98fb2d8648294a4a8da] Merge branch
>> 'for-next' of
>> https://git.kernel.org/pub/scm/linux/kernel/git/khilman/linux-omap.git
>> git bisect bad d154d0766bf3fe830e1bf98fb2d8648294a4a8da
>> # bad: [671b0fd929bec9ebf863cb376baf9b9d1226a510] Merge branch 'master'
>> of https://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux.git
>> git bisect bad 671b0fd929bec9ebf863cb376baf9b9d1226a510
>> # good: [f4fb100039e96211609dfc44fb24b9e4a8a0f2f9] Merge tag
>> 's390-7.2-4' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
>> git bisect good f4fb100039e96211609dfc44fb24b9e4a8a0f2f9
>> # good: [c296d513e034df93ddb230ad1c7a118d6b58750c] Merge branch
>> 'arm/fixes' of https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git
>> git bisect good c296d513e034df93ddb230ad1c7a118d6b58750c
>> # good: [de53c0a584f08f31eb957f8c849a59fb33b27fa8] Merge branch 'for-rc'
>> of https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git
>> git bisect good de53c0a584f08f31eb957f8c849a59fb33b27fa8
>> # bad: [a6969de5a038bc08a948ada52bd7199a8dfa6cec] Merge branch
>> 'usb-linus' of
>> https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb.git
>> git bisect bad a6969de5a038bc08a948ada52bd7199a8dfa6cec
>> # good: [15eab2dd63a97587405ab6bfaad9728d5b35b5fe] Merge tag
>> 'asoc-fix-v7.2-rc2' of
>> https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus
>> git bisect good 15eab2dd63a97587405ab6bfaad9728d5b35b5fe
>> # bad: [c2e819be6a5c7f34344926b4bd7e3dfca58cf48a] usb: gadget: printer:
>> fix infinite loop in printer_read()
>> git bisect bad c2e819be6a5c7f34344926b4bd7e3dfca58cf48a
>> # good: [29a142d3e8b35ebc9e0bcc78f4bc26c9b6a9ac0b] USB: gadget:
>> snps-udc: fix device name leak on probe failure
>> git bisect good 29a142d3e8b35ebc9e0bcc78f4bc26c9b6a9ac0b
>> # good: [1febec7e47cdcd01f43fb0211094e3010474666e] usb: gadget: f_ncm:
>> validate datagram bounds in ncm_unwrap_ntb()
>> git bisect good 1febec7e47cdcd01f43fb0211094e3010474666e
>> # bad: [0583f2fbf8f86ae3a0ce054f96783dd83e65d9bb] usb: gadget: udc: bdc:
>> free IRQ and drain func_wake_notify before teardown
>> git bisect bad 0583f2fbf8f86ae3a0ce054f96783dd83e65d9bb
>> # bad: [7aa7d4bf9d3fa9a6a47b640ad103ab433b7ff261] usb: typec: ucsi: Fix
>> race condition and ordering in port unregistration
>> git bisect bad 7aa7d4bf9d3fa9a6a47b640ad103ab433b7ff261
>> # first bad commit: [7aa7d4bf9d3fa9a6a47b640ad103ab433b7ff261] usb:
>> typec: ucsi: Fix race condition and ordering in port unregistration
>>>
>>> Cc: stable@vger.kernel.org
>>> Fixes: b9aa02ca39a4 ("usb: typec: ucsi: Add polling mechanism for partner tasks like alt mode checking")
>>> Fixes: b13abcb7ddd8 ("usb: typec: ucsi: Fix NULL pointer access")
>>> Fixes: fac4b8633fd6 ("usb: ucsi: Ensure connector delayed work items are flushed")
>>> Signed-off-by: Andrei Kuchynski <akuchynski@chromium.org>
>>> ---
>>>    drivers/usb/typec/ucsi/ucsi.c | 82 +++++++++++++++++------------------
>>>    1 file changed, 39 insertions(+), 43 deletions(-)
>>>
>>> diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c
>>> index 92166a3725b16..d9668ed7c80ea 100644
>>> --- a/drivers/usb/typec/ucsi/ucsi.c
>>> +++ b/drivers/usb/typec/ucsi/ucsi.c
>>> @@ -1845,6 +1845,42 @@ static int ucsi_register_port(struct ucsi *ucsi, struct ucsi_connector *con)
>>>        return ret;
>>>    }
>>>
>>> +static void ucsi_unregister_port(struct ucsi_connector *con)
>>> +{
>>> +     struct ucsi_work *uwork;
>>> +
>>> +     if (con->wq) {
>>> +             mutex_lock(&con->lock);
>>> +             ucsi_unregister_partner(con);
>>> +             /*
>>> +              * queue delayed items immediately so they can execute
>>> +              * and free themselves before the wq is destroyed
>>> +              */
>>> +             list_for_each_entry(uwork, &con->partner_tasks, node) {
>>> +                     if (cancel_delayed_work(&uwork->work))
>>> +                             queue_delayed_work(con->wq, &uwork->work, 0);
>>> +             }
>>> +             mutex_unlock(&con->lock);
>>> +
>>> +             destroy_workqueue(con->wq);
>>> +             con->wq = NULL;
>>> +     } else {
>>> +             ucsi_unregister_partner(con);
>>> +     }
>>> +
>>> +     ucsi_unregister_altmodes(con, UCSI_RECIPIENT_CON);
>>> +     ucsi_unregister_port_psy(con);
>>> +
>>> +     usb_power_delivery_unregister_capabilities(con->port_sink_caps);
>>> +     con->port_sink_caps = NULL;
>>> +     usb_power_delivery_unregister_capabilities(con->port_source_caps);
>>> +     con->port_source_caps = NULL;
>>> +     usb_power_delivery_unregister(con->pd);
>>> +     con->pd = NULL;
>>> +     typec_unregister_port(con->port);
>>> +     con->port = NULL;
>>> +}
>>> +
>>>    static u64 ucsi_get_supported_notifications(struct ucsi *ucsi)
>>>    {
>>>        u16 features = ucsi->cap.features;
>>> @@ -1971,22 +2007,8 @@ static int ucsi_init(struct ucsi *ucsi)
>>>        for (i = 0; i < ucsi->cap.num_connectors; i++)
>>>                lockdep_unregister_key(&connector[i].lock_key);
>>>
>>> -     for (con = connector; con->port; con++) {
>>> -             if (con->wq)
>>> -                     destroy_workqueue(con->wq);
>>> -             ucsi_unregister_partner(con);
>>> -             ucsi_unregister_altmodes(con, UCSI_RECIPIENT_CON);
>>> -             ucsi_unregister_port_psy(con);
>>> -
>>> -             usb_power_delivery_unregister_capabilities(con->port_sink_caps);
>>> -             con->port_sink_caps = NULL;
>>> -             usb_power_delivery_unregister_capabilities(con->port_source_caps);
>>> -             con->port_source_caps = NULL;
>>> -             usb_power_delivery_unregister(con->pd);
>>> -             con->pd = NULL;
>>> -             typec_unregister_port(con->port);
>>> -             con->port = NULL;
>>> -     }
>>> +     for (con = connector; con->port; con++)
>>> +             ucsi_unregister_port(con);
> 
> Hi Chaitanya,
> Thank you for spotting this!
> 
> Could you verify if calling lockdep_unregister_key(&connector[i].lock_key)
> after ucsi_unregister_port(con) resolves the issue?
> 
>       for (con = connector; con->port; con++)
>              ucsi_unregister_port(con);
>       for (i = 0; i < ucsi->cap.num_connectors; i++)
>                 lockdep_unregister_key(&connector[i].lock_key);
> 
> I will also try testing this on my side.

Thank you Andrei for the patch. It seems to work for us.

==
Chaitanya

> 
> Thanks,
> Andrei