[PATCH 4/5] staging: vchiq_arm: Create keep-alive thread during probe

[Maíra Canal]

Hi Stefan,

On 26/06/25 17:14, Stefan Wahren wrote:
> Hi Maíra,
> 
> [add Arnd]
> 
> Am 26.06.25 um 20:22 schrieb Maíra Canal:
>> Hi Stefan,
>>
>> On 09/03/25 09:50, Stefan Wahren wrote:
>>> Creating the keep-alive thread in vchiq_platform_init_state have
>>> the following advantages:
>>> - abort driver probe if kthread_create fails (more consistent behavior)
>>> - make resource release process easier
>>>
>>> Since vchiq_keepalive_thread_func is defined below
>>> vchiq_platform_init_state, the latter must be moved.
>>>
>>> Signed-off-by: Stefan Wahren <wahrenst at gmx.net>
>>> ---
>>>   .../interface/vchiq_arm/vchiq_arm.c           | 69 +++++++++----------
>>>   1 file changed, 34 insertions(+), 35 deletions(-)
>>>
>>
>> After this patch landed on 6.12 stable, I started to observe the
>> following warning in Mesa CI [1]:
> Is this also reproducible with 6.15.x or mainline ?

I was able to reproduce it with 6.16.0-rc2 [2].

[2] https://gitlab.freedesktop.org/mairacanal/mesa/-/jobs/79092300

Best Regards,
- Maíra

>>
>> 20:07:07.830: [  242.653532] INFO: task vchiq-keep/0:85 blocked for 
>> more than 120 seconds.
>> 20:07:07.835: [  242.660404]       Not tainted 6.12.34-v8-+ #13
>> 20:07:07.843: [  242.666173] "echo 0 > /proc/sys/kernel/ 
>> hung_task_timeout_secs" disables this message.
>> 20:07:07.852: [  242.677126] task:vchiq-keep/0    state:D stack:0 
>> pid:85    tgid:85    ppid:2      flags:0x00000008
>> 20:07:07.854: [  242.690734] Call trace:
>> 20:07:07.857: [  242.693191]  __switch_to+0x188/0x230
>> 20:07:07.858: [  242.697010]  __schedule+0xa54/0xb28
>> 20:07:07.859: [  242.704889]  schedule+0x80/0x120
>> 20:07:07.859: [  242.712581]  schedule_preempt_disabled+0x30/0x50
>> 20:07:07.860: [  242.721437]  kthread+0xd4/0x1a0
>> 20:07:07.861: [  242.724606]  ret_from_fork+0x10/0x20
>>
>> If I revert this patch, I no longer get those warnings. From that, I was
>> wondering: is it possible that in some scenarios, we might never
>> actually get to the connected state and therefore, we wouldn't wake up
>> the kthread that we created when we initialized the state?
> Sure we communicate to the VideoCore and there might be reasons the 
> connection fail. I'm not aware that a missing wakeup cause a hanging task.
> 
> Best regards
>>
>> [1] https://gitlab.freedesktop.org/mairacanal/mesa/-/jobs/78940369
>>
>> Best Regards,
>> - Maíra
>>
>>> diff --git a/drivers/staging/vc04_services/interface/vchiq_arm/ 
>>> vchiq_arm.c b/drivers/staging/vc04_services/interface/vchiq_arm/ 
>>> vchiq_arm.c
>>> index 0c7ea2d0ee85..64f9536f1232 100644
>>> --- a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c
>>> +++ b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c
>>> @@ -280,29 +280,6 @@ static int vchiq_platform_init(struct 
>>> platform_device *pdev, struct vchiq_state
>>>       return 0;
>>>   }
>>>
>>> -int
>>> -vchiq_platform_init_state(struct vchiq_state *state)
>>> -{
>>> -    struct vchiq_arm_state *platform_state;
>>> -
>>> -    platform_state = devm_kzalloc(state->dev, 
>>> sizeof(*platform_state), GFP_KERNEL);
>>> -    if (!platform_state)
>>> -        return -ENOMEM;
>>> -
>>> -    rwlock_init(&platform_state->susp_res_lock);
>>> -
>>> -    init_completion(&platform_state->ka_evt);
>>> -    atomic_set(&platform_state->ka_use_count, 0);
>>> -    atomic_set(&platform_state->ka_use_ack_count, 0);
>>> -    atomic_set(&platform_state->ka_release_count, 0);
>>> -
>>> -    platform_state->state = state;
>>> -
>>> -    state->platform_state = (struct opaque_platform_state 
>>> *)platform_state;
>>> -
>>> -    return 0;
>>> -}
>>> -
>>>   static struct vchiq_arm_state *vchiq_platform_get_arm_state(struct 
>>> vchiq_state *state)
>>>   {
>>>       return (struct vchiq_arm_state *)state->platform_state;
>>> @@ -1011,6 +988,39 @@ vchiq_keepalive_thread_func(void *v)
>>>       return 0;
>>>   }
>>>
>>> +int
>>> +vchiq_platform_init_state(struct vchiq_state *state)
>>> +{
>>> +    struct vchiq_arm_state *platform_state;
>>> +    char threadname[16];
>>> +
>>> +    platform_state = devm_kzalloc(state->dev, 
>>> sizeof(*platform_state), GFP_KERNEL);
>>> +    if (!platform_state)
>>> +        return -ENOMEM;
>>> +
>>> +    snprintf(threadname, sizeof(threadname), "vchiq-keep/%d",
>>> +         state->id);
>>> +    platform_state->ka_thread = 
>>> kthread_create(&vchiq_keepalive_thread_func,
>>> +                           (void *)state, threadname);
>>> +    if (IS_ERR(platform_state->ka_thread)) {
>>> +        dev_err(state->dev, "couldn't create thread %s\n", threadname);
>>> +        return PTR_ERR(platform_state->ka_thread);
>>> +    }
>>> +
>>> +    rwlock_init(&platform_state->susp_res_lock);
>>> +
>>> +    init_completion(&platform_state->ka_evt);
>>> +    atomic_set(&platform_state->ka_use_count, 0);
>>> +    atomic_set(&platform_state->ka_use_ack_count, 0);
>>> +    atomic_set(&platform_state->ka_release_count, 0);
>>> +
>>> +    platform_state->state = state;
>>> +
>>> +    state->platform_state = (struct opaque_platform_state 
>>> *)platform_state;
>>> +
>>> +    return 0;
>>> +}
>>> +
>>>   int
>>>   vchiq_use_internal(struct vchiq_state *state, struct vchiq_service 
>>> *service,
>>>              enum USE_TYPE_E use_type)
>>> @@ -1331,7 +1341,6 @@ void vchiq_platform_conn_state_changed(struct 
>>> vchiq_state *state,
>>>                          enum vchiq_connstate newstate)
>>>   {
>>>       struct vchiq_arm_state *arm_state = 
>>> vchiq_platform_get_arm_state(state);
>>> -    char threadname[16];
>>>
>>>       dev_dbg(state->dev, "suspend: %d: %s->%s\n",
>>>           state->id, get_conn_state_name(oldstate), 
>>> get_conn_state_name(newstate));
>>> @@ -1346,17 +1355,7 @@ void vchiq_platform_conn_state_changed(struct 
>>> vchiq_state *state,
>>>
>>>       arm_state->first_connect = 1;
>>>       write_unlock_bh(&arm_state->susp_res_lock);
>>> -    snprintf(threadname, sizeof(threadname), "vchiq-keep/%d",
>>> -         state->id);
>>> -    arm_state->ka_thread = kthread_create(&vchiq_keepalive_thread_func,
>>> -                          (void *)state,
>>> -                          threadname);
>>> -    if (IS_ERR(arm_state->ka_thread)) {
>>> -        dev_err(state->dev, "suspend: Couldn't create thread %s\n",
>>> -            threadname);
>>> -    } else {
>>> -        wake_up_process(arm_state->ka_thread);
>>> -    }
>>> +    wake_up_process(arm_state->ka_thread);
>>>   }
>>>
>>>   static const struct of_device_id vchiq_of_match[] = {
>>> -- 
>>> 2.34.1
>>>
>>>
>>
>>
>