Re: [PATCH v4 drm-dp 02/11] drm/hisilicon/hibmc: fix dp probabilistical detect errors after HPD irq

[Yongbang Shi <shiyongbang@huawei.com>]

> On Fri, Sep 12, 2025 at 09:23:05AM +0800, Yongbang Shi wrote:
>>> On Thu, Sep 11, 2025 at 05:32:40PM +0800, Yongbang Shi wrote:
>>>>> On Thu, Aug 14, 2025 at 08:19:41PM +0800, Yongbang Shi wrote:
>>>>>>> On Wed, Aug 13, 2025 at 05:42:29PM +0800, Yongbang Shi wrote:
>>>>>>>> From: Baihan Li <libaihan@huawei.com>
>>>>>>>>
>>>>>>>> The debouncing when HPD pulled out still remains sometimes, 200ms still can
>>>>>>>> not ensure helper_detect() is correct. So add a flag to hold the sink
>>>>>>>> status, and changed detect_ctx() functions by using flag to check status.
>>>>>>> THis doesn't explain what is wrong with
>>>>>>> drm_connector_helper_detect_from_ddc(). In the end, this function
>>>>>>> doesn't use the HPD pin.
>>>>>> I'm sorry about the misunderstanding.
>>>>>> The issue is that after plugging or unplugging the monitor, the driver takes no action sometimes
>>>>>> even though an interrupt is triggered. The root cause is that drm_connector_helper_detect_from_ddc()
>>>>>> still returns connected status when the monitor is unplugged.
>>>>>> And I will fix the way in the end.
>>>>> Can you perform a normal DP detection: read DPCD and check that there is
>>>>> a DPRX attached and that it's either non-branch device or it has one or
>>>>> more sinks?
>>>> I'm very sorry that I didn't get the last sentence's asking before.
>>>> It's a non-branch device. We just connect a DP monitor.
>>> Somebody might connect a different configuration than the one that you
>>> are using.
>> Okay, I can add the check drm_dp_is_branch() in the DP's detect_ctx() to
>> intercept branch devices, is that good?
> My suggestion is to implement DP detection in the way it's done by other
> DP drivers.

Okay, I will reference the code from other manufacturer ways.

Thanks,
Baihan.