[PATCH v8 00/18] Apply drm_bridge_connector and panel_bridge helper for the Analogix DP driver

[Damon Ding]

Hi Heiko,

On 1/20/2026 12:06 AM, Heiko Stübner wrote:
> Hi Damon,
> 
> Am Mittwoch, 17. Dezember 2025, 10:33:03 Mitteleuropäische Normalzeit schrieb Damon Ding:
>> PATCH 1 is to ensure the last bridge declares OP_EDID or OP_MODES or both.
>> PATCH 2 is a small format optimization for struct analogid_dp_device.
>> PATCH 3 is to perform mode setting in &drm_bridge_funcs.atomic_enable.
>> PATCH 4 is to add a new parameter to store the point of next bridge.
>> PATCH 5 is to make legacy bridge driver more universal.
>> PATCH 6-11 are preparations for apply drm_bridge_connector helper.
>> PATCH 12 is to apply the drm_bridge_connector helper.
>> PATCH 13-15 are to move the panel/bridge parsing to the Analogix side.
>> PATCH 16 is to attach the next bridge on Analogix side uniformly.
>> PATCH 17-18 are to apply the panel_bridge helper.
> 
> On top of all the recent bridge-related commits, this series still works
> on my rk3588-eDP board - when the display is plugged in on boot.
> 
> As that is the current state we also had before, this series:
> Tested-by: Heiko Stuebner <heiko at sntech.de> on rk3588
> 
> 
> ======
> !!! The following should not hold up this series though, because the
> analogix-dp not handling hotplug, is also the current state already!!!
> ======
> 
> However, actually hotplugging the device does not seem to work.
> 
> With the hpd-gpio inside a dp-connector node.
> 
> 
> (1) When booting connected and unplugging the display I get:
> 
> [   42.725888] rockchip-dp fdec0000.edp: Rx Max Link Rate is abnormal :0 !
> [   42.733325] rockchip-dp fdec0000.edp: Rx Max Lane count is abnormal :0 !
> [   42.764853] rockchip-dp fdec0000.edp: LT link start failed!
> [   42.771105] rockchip-dp fdec0000.edp: eDP link training failed (-110)
> [   42.778329] rockchip-dp fdec0000.edp: unable to do link train, ret=-110
> [   42.785747] [drm:analogix_dp_bridge_atomic_enable] *ERROR* dp commit error, ret = -110
> ( ... repeasts a bunch of times ... )
> [   43.284897] rockchip-dp fdec0000.edp: failed to set bridge, retry: 4
> [   43.292045] rockchip-dp fdec0000.edp: too many times retry set bridge, give it up
> 
> plugging the eDP back in reenables the display successfully though.
> 
> (2) Booting with the display unplugged
> 
> Same messages as above, but plugging the display in for the first time
> does not create an output.
> 
> I'll try to dig more, but if you have an idea, what I should look at, I'd,
> be really grateful.
> 

Sorry for the late reply. I've been heavily occupied with the downstream 
work recently.

I can reproduce the same issue on my RK3576 IOTEST board, which is 
equipped with a standard DP receptacle for eDP interface.

(BTW: The RK3576 Analogix DP functions properly with a set of targeted 
changes, and I plan to upstream these modifications later. :-))

Regarding the eDP HPD exception, several functional patches for HPD 
handling have not yet been synced to the upstream. I will verify these 
changes and submit them when time permits.

Best regards,
Damon