Re: [PATCH v2 09/10] drm/bridge: analogix_dp: Restrict forced connected status only for panel endpoint

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] Missing early return in analogix_dp_bridge_detect() causes unintended hardware accesses and significant delays for panel bridges.
- [Medium] Missing drm_bridge_put() causes a reference count leak for the bridge returned by drm_bridge_chain_get_last_bridge().
--

--- Patch [9]: [PATCH v2 09/10] drm/bridge: analogix_dp: Restrict forced connected status only for panel endpoint ---
commit 7e658941bfdab69a58f5d04d4acd4ae671e37ea8
Author: Damon Ding <[email protected]>

drm/bridge: analogix_dp: Restrict forced connected status only for panel endpoint

This commit limits the unconditional connected status shortcut to cases
where the last bridge in the chain is a panel. For other cases, it relies
on analogix_dp_detect_hpd() to report accurate hotplug state.

Link: https://lore.kernel.org/all/5209243.oAaWjsssbx@diego/

> diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
> index 1c85ca93ee7b1..f7db5b5a3f917 100644
> --- a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
> +++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
> @@ -910,8 +910,15 @@ analogix_dp_bridge_detect(struct drm_bridge *bridge, struct drm_connector *conne
>  	struct analogix_dp_device *dp = to_dp(bridge);
>  	enum drm_connector_status status = connector_status_disconnected;
>  
> -	if (dp->plat_data->next_bridge)
> -		return connector_status_connected;
> +	/*
> +	 * Only assume connected when the bridge chain ends with a panel.
> +	 * Panel bridges lack a detect callback and historically rely on
> +	 * this shortcut. For other next bridge types (e.g., DP connector
> +	 * bridges), fall through to analogix_dp_detect_hpd() to report
> +	 * accurate hotplug state.
> +	 */
> +	if (dp->plat_data->next_bridge && dp->last_bridge_is_panel)
> +		status = connector_status_connected;
>  
>  	if (!analogix_dp_detect_hpd(dp))
>  		status = connector_status_connected;

[Severity: High]
Does this unintentionally fall through and execute analogix_dp_detect_hpd()
even when the bridge chain ends with a panel?

Previously, analogix_dp_bridge_detect() returned early here. By just
assigning the status without returning, it looks like panel bridges will now
unconditionally execute analogix_dp_detect_hpd(). This can introduce an
unnecessary 100ms delay and unintended register/AUX writes to manually force
HPD on hardware that lacks an HPD signal.

Should there be an early return here instead of continuing execution?

[ ... ]

> @@ -938,6 +946,10 @@ static int analogix_dp_bridge_attach(struct drm_bridge *bridge,
>  			dev_err(dp->dev, "failed to attach following panel or bridge (%d)\n", ret);
>  			return ret;
>  		}
> +
> +		last_bridge = drm_bridge_chain_get_last_bridge(dp->encoder);
> +		if (drm_bridge_is_panel(last_bridge))
> +			dp->last_bridge_is_panel = true;
>  	}
>  
>  	return 0;

[Severity: Medium]
Does this leak the reference to the last bridge in analogix_dp_bridge_attach()?

drm_bridge_chain_get_last_bridge() increments the reference count of the
returned bridge, but it doesn't appear that drm_bridge_put() is called
before the function returns. Will this prevent proper cleanup when the
device is unbound or the module is unloaded?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=9