Hi Dmitry,
On 1/13/25 11:16 AM, Dmitry Baryshkov wrote:
> On Sat, Jan 11, 2025 at 12:04:09AM +0200, Cristian Ciocaltea wrote:
>> The recently introduced hotplug event handler in the HDMI Connector
>> framework attempts to unconditionally read the EDID data, leading to a
>> bunch of non-harmful, yet quite annoying DDC/I2C related errors being
>> reported.
>>
>> Ensure the operation is performed only for connectors having the status
>> connected or unknown.
>>
>> Fixes: ab716b74dc9d ("drm/display/hdmi: implement hotplug functions")
>> Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@xxxxxxxxxxxxx>
>> ---
>> drivers/gpu/drm/display/drm_hdmi_state_helper.c | 2 ++
>> 1 file changed, 2 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/display/drm_hdmi_state_helper.c b/drivers/gpu/drm/display/drm_hdmi_state_helper.c
>> index 2691e8b3e480131ac6e4e4b74b24947be55694bd..8e4b30e09b53b84cfd36199d56db3221a00085b0 100644
>> --- a/drivers/gpu/drm/display/drm_hdmi_state_helper.c
>> +++ b/drivers/gpu/drm/display/drm_hdmi_state_helper.c
>> @@ -786,8 +786,10 @@ drm_atomic_helper_connector_hdmi_update(struct drm_connector *connector,
>> const struct drm_edid *drm_edid;
>>
>> if (status == connector_status_disconnected) {
>> + drm_edid_connector_update(connector, NULL);
>> // TODO: also handle CEC and scramber, HDMI sink disconnected.
>> drm_connector_hdmi_audio_plugged_notify(connector, false);
>> + return;
>
> I think, it should be other way around: plugged_notify before
> drm_edid_connector_update(). At least that would follow current logic of
> the function.
Yeah, I wasn't really sure about the order here. Will get this fixed in v2.
Thanks,
Cristian
