Hi,

On 10/10/2025 10:02 PM, Dmitry Baryshkov wrote:
On Thu, Oct 09, 2025 at 12:10:42PM +0800, Damon Ding wrote:
Hi Luca,

On 10/2/2025 12:09 AM, Luca Ceresoli wrote:
Hello Damon,

On Tue, 30 Sep 2025 17:09:13 +0800
Damon Ding <[email protected]> wrote:

When multiple bridges are present, EDID detection capability
(DRM_BRIDGE_OP_EDID) takes precedence over modes detection
(DRM_BRIDGE_OP_MODES). To ensure the above two capabilities are
determined by the last bridge in the chain, we handle three cases:

Case 1: The later bridge declares only DRM_BRIDGE_OP_MODES
   - If the previous bridge declares DRM_BRIDGE_OP_EDID, set
     &drm_bridge_connector.bridge_edid to NULL and set
     &drm_bridge_connector.bridge_modes to the later bridge.
   - Ensure modes detection capability of the later bridge will not
     be ignored.

Case 2: The later bridge declares only DRM_BRIDGE_OP_EDID
   - If the previous bridge declares DRM_BRIDGE_OP_MODES, set
     &drm_bridge_connector.bridge_modes to NULL and set
     &drm_bridge_connector.bridge_edid to the later bridge.
   - Although EDID detection capability has higher priority, this
     operation is for balance and makes sense.

Case 3: the later bridge declares both of them
   - Assign later bridge as &drm_bridge_connector.bridge_edid and
     and &drm_bridge_connector.bridge_modes to this bridge.
   - Just leave transfer of these two capabilities as before.

I think the whole explanation can be more concisely rewritten as:

If the later bridge declares OP_EDID, OP_MODES or both, then both
.bridge_modes and .bridge_edid should be set to NULL (if any was set
from a previous bridge), and then .bridge_modes and/or .bridge_edid be
set to the later bridge as is done already.

Does this look correct (i.e. does it convey the same meaning)?

--- a/drivers/gpu/drm/display/drm_bridge_connector.c
+++ b/drivers/gpu/drm/display/drm_bridge_connector.c
@@ -640,6 +640,7 @@ struct drm_connector *drm_bridge_connector_init(struct 
drm_device *drm,
        struct drm_connector *connector;
        struct i2c_adapter *ddc = NULL;
        struct drm_bridge *bridge, *panel_bridge = NULL;
+       struct drm_bridge *pre_bridge_edid, *pre_bridge_modes;
        unsigned int supported_formats = BIT(HDMI_COLORSPACE_RGB);
        unsigned int max_bpc = 8;
        bool support_hdcp = false;
@@ -668,6 +669,9 @@ struct drm_connector *drm_bridge_connector_init(struct 
drm_device *drm,
         */
        connector_type = DRM_MODE_CONNECTOR_Unknown;
        drm_for_each_bridge_in_chain(encoder, bridge) {
+               pre_bridge_edid = bridge_connector->bridge_edid;
+               pre_bridge_modes = bridge_connector->bridge_modes;
+
                if (!bridge->interlace_allowed)
                        connector->interlace_allowed = false;
                if (!bridge->ycbcr_420_allowed)
@@ -681,6 +685,44 @@ struct drm_connector *drm_bridge_connector_init(struct 
drm_device *drm,
                        bridge_connector->bridge_detect = bridge;
                if (bridge->ops & DRM_BRIDGE_OP_MODES)
                        bridge_connector->bridge_modes = bridge;
+
+               /*
+                * When multiple bridges are present, EDID detection capability
+                * (DRM_BRIDGE_OP_EDID) takes precedence over modes detection
+                * (DRM_BRIDGE_OP_MODES). To ensure the above two capabilities
+                * are determined by the last bridge in the chain, we handle
+                * three cases:
+                *
+                * Case 1: The later bridge declares only DRM_BRIDGE_OP_MODES
+                *  - If the previous bridge declares DRM_BRIDGE_OP_EDID, set
+                *    &drm_bridge_connector.bridge_edid to NULL and set
+                *    &drm_bridge_connector.bridge_modes to the later bridge.
+                *  - Ensure modes detection capability of the later bridge
+                *    will not be ignored.
+                *
+                * Case 2: The later bridge declares only DRM_BRIDGE_OP_EDID
+                *  - If the previous bridge declares DRM_BRIDGE_OP_MODES, set
+                *    &drm_bridge_connector.bridge_modes to NULL and set
+                *    &drm_bridge_connector.bridge_edid to the later bridge.
+                *  - Although EDID detection capability has higher priority,
+                *    this operation is for balance and makes sense.
+                *
+                * Case 3: the later bridge declares both of them
+                *  - Assign later bridge as &drm_bridge_connector.bridge_edid
+                *    and &drm_bridge_connector.bridge_modes to this bridge.
+                *  - Just leave transfer of these two capabilities as before.
+                */
+               if (bridge->ops & DRM_BRIDGE_OP_EDID &&
+                   !(bridge->ops & DRM_BRIDGE_OP_MODES)) {
+                       if (pre_bridge_modes)
+                               bridge_connector->bridge_modes = NULL;
+               }
+               if (bridge->ops & DRM_BRIDGE_OP_MODES &&
+                   !(bridge->ops & DRM_BRIDGE_OP_EDID)) {
+                       if (pre_bridge_edid)
+                               bridge_connector->bridge_edid = NULL;
+               }
+

If the above rewrite is correct, then I think this patch can be
rewritten in a simple way (build tested only):

diff --git a/drivers/gpu/drm/display/drm_bridge_connector.c 
b/drivers/gpu/drm/display/drm_bridge_connector.c
index a5bdd6c10643..bd5dbafe88bc 100644
--- a/drivers/gpu/drm/display/drm_bridge_connector.c
+++ b/drivers/gpu/drm/display/drm_bridge_connector.c
@@ -672,14 +672,18 @@ struct drm_connector *drm_bridge_connector_init(struct 
drm_device *drm,
                  if (!bridge->ycbcr_420_allowed)
                          connector->ycbcr_420_allowed = false;
-               if (bridge->ops & DRM_BRIDGE_OP_EDID)
-                       bridge_connector->bridge_edid = bridge;
+               if (bridge->ops & DRM_BRIDGE_OP_EDID || bridge->ops & 
DRM_BRIDGE_OP_MODES) {
+                       bridge_connector->bridge_edid = NULL;
+                       bridge_connector->bridge_modes = NULL;
+                       if (bridge->ops & DRM_BRIDGE_OP_EDID)
+                               bridge_connector->bridge_edid = bridge;
+                       if (bridge->ops & DRM_BRIDGE_OP_MODES)
+                               bridge_connector->bridge_modes = bridge;
+               }
                  if (bridge->ops & DRM_BRIDGE_OP_HPD)
                          bridge_connector->bridge_hpd = bridge;
                  if (bridge->ops & DRM_BRIDGE_OP_DETECT)
                          bridge_connector->bridge_detect = bridge;
-               if (bridge->ops & DRM_BRIDGE_OP_MODES)
-                       bridge_connector->bridge_modes = bridge;
                  if (bridge->ops & DRM_BRIDGE_OP_HDMI) {
                          if (bridge_connector->bridge_hdmi)
                                  return ERR_PTR(-EBUSY);


Yes, this is correct and maintains functional equivalence with the previous
implementation.

I previously attempted to implement this feature by modifying the logic in
this section. However, that approach would obscure the explicit propagation
semantics of the bridge chain flags (OP_EDID/OP_HPD/OP_DETECT/OP_MODES).
Therefore, I finally decided to implemented it as a specific check after
this code block.

Dmitry, what's your take on this?

I think I prefer Luca's code, it is simpler and easier to understand. It
doesn't need a huge comment, something like "leave the last bridge which
provides either OP_EDID or OP_MODES" should be enough.


Yes, I will update the code in v7.


Another thing to note is that this patch conflicts with [0], which I
plan to apply in the next few days. The two patches are orthogonal but
they insist on the same lines (those assigning
bridge_connector->bridge_* = bridge). Not a big deal, whichever patch
comes later will be easily adapted. Just wanted to ensure you are aware.

[0] 
https://lore.kernel.org/all/20250926-drm-bridge-alloc-getput-bridge-connector-v2-1-138b4bb70...@bootlin.com/


This is indeed a clever approach to the managing bridge resource cleanup in
drm_bridge_connector. Thanks a lot for the heads-up! I'll resolve this
conflict and rebase the patch series.

Apologies for the delayed reply as I was on vacation. ;-)

Best regards,
Damon

Reply via email to