[PATCH v2] drm: of: Properly try all possible cases for bridge/panel detection

[PATCH v2] drm: of: Properly try all possible cases for bridge/panel detection

[Paul Kocialkowski]

While bridge/panel detection was initially relying on the usual
port/ports-based of graph detection, it was recently changed to
perform the lookup on any child node that is not port/ports
instead when such a node is available, with no fallback on the
usual way.

This results in breaking detection when a child node is present
but does not contain any panel or bridge node, even when the
usual port/ports-based of graph is there.

In order to support both situations properly, this commit reworks
the logic to try both options and not just one of the two: it will
only return -EPROBE_DEFER when both have failed.

Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Fixes: 80253168dbfd ("drm: of: Lookup if child node has panel or bridge")
---

Changes since v1:
- Renamed remote to node;
- Renamed helper to find_panel_or_bridge;
- Cleared bridge pointer early;
- Returned early to make the code more concise;

---
 drivers/gpu/drm/drm_of.c | 98 ++++++++++++++++++++--------------------
 1 file changed, 49 insertions(+), 49 deletions(-)

diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
index 9d90cd75c457..63137c833b7a 100644
--- a/drivers/gpu/drm/drm_of.c
+++ b/drivers/gpu/drm/drm_of.c
@@ -219,6 +219,28 @@ int drm_of_encoder_active_endpoint(struct device_node *node,
 }
 EXPORT_SYMBOL_GPL(drm_of_encoder_active_endpoint);
 
+static int find_panel_or_bridge(struct device_node *node,
+				struct drm_panel **panel,
+				struct drm_bridge **bridge)
+{
+	if (panel) {
+		*panel = of_drm_find_panel(node);
+		if (!IS_ERR(*panel))
+			return 0;
+		else
+			*panel = NULL;
+	}
+
+	/* No panel found yet, check for a bridge next. */
+	if (bridge) {
+		*bridge = of_drm_find_bridge(node);
+		if (*bridge)
+			return 0;
+	}
+
+	return -EPROBE_DEFER;
+}
+
 /**
  * drm_of_find_panel_or_bridge - return connected panel or bridge device
  * @np: device tree node containing encoder output ports
@@ -241,66 +263,44 @@ int drm_of_find_panel_or_bridge(const struct device_node *np,
 				struct drm_panel **panel,
 				struct drm_bridge **bridge)
 {
-	int ret = -EPROBE_DEFER;
-	struct device_node *remote;
+	struct device_node *node;
+	int ret;
 
 	if (!panel && !bridge)
 		return -EINVAL;
+
 	if (panel)
 		*panel = NULL;
-
-	/**
-	 * Devices can also be child nodes when we also control that device
-	 * through the upstream device (ie, MIPI-DCS for a MIPI-DSI device).
-	 *
-	 * Lookup for a child node of the given parent that isn't either port
-	 * or ports.
-	 */
-	for_each_available_child_of_node(np, remote) {
-		if (of_node_name_eq(remote, "port") ||
-		    of_node_name_eq(remote, "ports"))
-			continue;
-
-		goto of_find_panel_or_bridge;
+	if (bridge)
+		*bridge = NULL;
+
+	/* Check for a graph on the device node first. */
+	if (of_graph_is_present(np)) {
+		node = of_graph_get_remote_node(np, port, endpoint);
+		if (node) {
+			ret = find_panel_or_bridge(node, panel, bridge);
+			of_node_put(node);
+
+			if (!ret)
+				return 0;
+		}
 	}
 
-	/*
-	 * of_graph_get_remote_node() produces a noisy error message if port
-	 * node isn't found and the absence of the port is a legit case here,
-	 * so at first we silently check whether graph presents in the
-	 * device-tree node.
-	 */
-	if (!of_graph_is_present(np))
-		return -ENODEV;
-
-	remote = of_graph_get_remote_node(np, port, endpoint);
-
-of_find_panel_or_bridge:
-	if (!remote)
-		return -ENODEV;
-
-	if (panel) {
-		*panel = of_drm_find_panel(remote);
-		if (!IS_ERR(*panel))
-			ret = 0;
-		else
-			*panel = NULL;
-	}
+	/* Otherwise check for any child node other than port/ports. */
+	for_each_available_child_of_node(np, node) {
+		if (of_node_name_eq(node, "port") ||
+		    of_node_name_eq(node, "ports"))
+			continue;
 
-	/* No panel found yet, check for a bridge next. */
-	if (bridge) {
-		if (ret) {
-			*bridge = of_drm_find_bridge(remote);
-			if (*bridge)
-				ret = 0;
-		} else {
-			*bridge = NULL;
-		}
+		ret = find_panel_or_bridge(node, panel, bridge);
+		of_node_put(node);
 
+		/* Stop at the first found occurrence. */
+		if (!ret)
+			return 0;
 	}
 
-	of_node_put(remote);
-	return ret;
+	return -EPROBE_DEFER;
 }
 EXPORT_SYMBOL_GPL(drm_of_find_panel_or_bridge);
 
-- 
2.35.1

Re: [PATCH v2] drm: of: Properly try all possible cases for bridge/panel detection

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 1946 bytes --]

On Fri, Mar 18, 2022 at 05:02:49PM +0100, Paul Kocialkowski wrote:
> While bridge/panel detection was initially relying on the usual
> port/ports-based of graph detection, it was recently changed to
> perform the lookup on any child node that is not port/ports
> instead when such a node is available, with no fallback on the
> usual way.
> 
> This results in breaking detection when a child node is present
> but does not contain any panel or bridge node, even when the
> usual port/ports-based of graph is there.
> 
> In order to support both situations properly, this commit reworks
> the logic to try both options and not just one of the two: it will
> only return -EPROBE_DEFER when both have failed.
> 
> Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
> Fixes: 80253168dbfd ("drm: of: Lookup if child node has panel or bridge")
> ---
> 
> Changes since v1:
> - Renamed remote to node;
> - Renamed helper to find_panel_or_bridge;
> - Cleared bridge pointer early;
> - Returned early to make the code more concise;
> 
> ---
>  drivers/gpu/drm/drm_of.c | 98 ++++++++++++++++++++--------------------
>  1 file changed, 49 insertions(+), 49 deletions(-)
> 
> diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
> index 9d90cd75c457..63137c833b7a 100644
> --- a/drivers/gpu/drm/drm_of.c
> +++ b/drivers/gpu/drm/drm_of.c
> @@ -219,6 +219,28 @@ int drm_of_encoder_active_endpoint(struct device_node *node,
>  }
>  EXPORT_SYMBOL_GPL(drm_of_encoder_active_endpoint);
>  
> +static int find_panel_or_bridge(struct device_node *node,
> +				struct drm_panel **panel,
> +				struct drm_bridge **bridge)
> +{
> +	if (panel) {
> +		*panel = of_drm_find_panel(node);
> +		if (!IS_ERR(*panel))
> +			return 0;
> +		else
> +			*panel = NULL;

You don't need the else branch here, we already cleared panel in
drm_of_find_panel_or_bridge

Looks good otherwise, thanks!
Maxime

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

Re: [PATCH v2] drm: of: Properly try all possible cases for bridge/panel detection

[Paul Kocialkowski]

[-- Attachment #1: Type: text/plain, Size: 2402 bytes --]

Hi Maxime,

On Fri 18 Mar 22, 17:18, Maxime Ripard wrote:
> On Fri, Mar 18, 2022 at 05:02:49PM +0100, Paul Kocialkowski wrote:
> > While bridge/panel detection was initially relying on the usual
> > port/ports-based of graph detection, it was recently changed to
> > perform the lookup on any child node that is not port/ports
> > instead when such a node is available, with no fallback on the
> > usual way.
> > 
> > This results in breaking detection when a child node is present
> > but does not contain any panel or bridge node, even when the
> > usual port/ports-based of graph is there.
> > 
> > In order to support both situations properly, this commit reworks
> > the logic to try both options and not just one of the two: it will
> > only return -EPROBE_DEFER when both have failed.
> > 
> > Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
> > Fixes: 80253168dbfd ("drm: of: Lookup if child node has panel or bridge")
> > ---
> > 
> > Changes since v1:
> > - Renamed remote to node;
> > - Renamed helper to find_panel_or_bridge;
> > - Cleared bridge pointer early;
> > - Returned early to make the code more concise;
> > 
> > ---
> >  drivers/gpu/drm/drm_of.c | 98 ++++++++++++++++++++--------------------
> >  1 file changed, 49 insertions(+), 49 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
> > index 9d90cd75c457..63137c833b7a 100644
> > --- a/drivers/gpu/drm/drm_of.c
> > +++ b/drivers/gpu/drm/drm_of.c
> > @@ -219,6 +219,28 @@ int drm_of_encoder_active_endpoint(struct device_node *node,
> >  }
> >  EXPORT_SYMBOL_GPL(drm_of_encoder_active_endpoint);
> >  
> > +static int find_panel_or_bridge(struct device_node *node,
> > +				struct drm_panel **panel,
> > +				struct drm_bridge **bridge)
> > +{
> > +	if (panel) {
> > +		*panel = of_drm_find_panel(node);
> > +		if (!IS_ERR(*panel))
> > +			return 0;
> > +		else
> > +			*panel = NULL;
> 
> You don't need the else branch here, we already cleared panel in
> drm_of_find_panel_or_bridge

I think we do, because of_drm_find_panel doesn't return NULL when an error
happens but PTR_ERR(), so we need to clear it to keep up with previous
expectations.

> Looks good otherwise, thanks!

Great, thanks for the review!

Paul

-- 
Paul Kocialkowski, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH v2] drm: of: Properly try all possible cases for bridge/panel detection

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 2482 bytes --]

On Fri, Mar 18, 2022 at 05:22:46PM +0100, Paul Kocialkowski wrote:
> On Fri 18 Mar 22, 17:18, Maxime Ripard wrote:
> > On Fri, Mar 18, 2022 at 05:02:49PM +0100, Paul Kocialkowski wrote:
> > > While bridge/panel detection was initially relying on the usual
> > > port/ports-based of graph detection, it was recently changed to
> > > perform the lookup on any child node that is not port/ports
> > > instead when such a node is available, with no fallback on the
> > > usual way.
> > > 
> > > This results in breaking detection when a child node is present
> > > but does not contain any panel or bridge node, even when the
> > > usual port/ports-based of graph is there.
> > > 
> > > In order to support both situations properly, this commit reworks
> > > the logic to try both options and not just one of the two: it will
> > > only return -EPROBE_DEFER when both have failed.
> > > 
> > > Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
> > > Fixes: 80253168dbfd ("drm: of: Lookup if child node has panel or bridge")
> > > ---
> > > 
> > > Changes since v1:
> > > - Renamed remote to node;
> > > - Renamed helper to find_panel_or_bridge;
> > > - Cleared bridge pointer early;
> > > - Returned early to make the code more concise;
> > > 
> > > ---
> > >  drivers/gpu/drm/drm_of.c | 98 ++++++++++++++++++++--------------------
> > >  1 file changed, 49 insertions(+), 49 deletions(-)
> > > 
> > > diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
> > > index 9d90cd75c457..63137c833b7a 100644
> > > --- a/drivers/gpu/drm/drm_of.c
> > > +++ b/drivers/gpu/drm/drm_of.c
> > > @@ -219,6 +219,28 @@ int drm_of_encoder_active_endpoint(struct device_node *node,
> > >  }
> > >  EXPORT_SYMBOL_GPL(drm_of_encoder_active_endpoint);
> > >  
> > > +static int find_panel_or_bridge(struct device_node *node,
> > > +				struct drm_panel **panel,
> > > +				struct drm_bridge **bridge)
> > > +{
> > > +	if (panel) {
> > > +		*panel = of_drm_find_panel(node);
> > > +		if (!IS_ERR(*panel))
> > > +			return 0;
> > > +		else
> > > +			*panel = NULL;
> > 
> > You don't need the else branch here, we already cleared panel in
> > drm_of_find_panel_or_bridge
> 
> I think we do, because of_drm_find_panel doesn't return NULL when an error
> happens but PTR_ERR(), so we need to clear it to keep up with previous
> expectations.

Ah, right. We should remove the else then, checkpatch complains about it.

Maxime

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]