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 <[email protected]>
Fixes: 80253168dbfd ("drm: of: Lookup if child node has panel or bridge")
---
drivers/gpu/drm/drm_of.c | 93 +++++++++++++++++++++-------------------
1 file changed, 49 insertions(+), 44 deletions(-)
diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
index 9d90cd75c457..67f1b7dfc892 100644
--- a/drivers/gpu/drm/drm_of.c
+++ b/drivers/gpu/drm/drm_of.c
@@ -219,6 +219,35 @@ int drm_of_encoder_active_endpoint(struct device_node *node,
}
EXPORT_SYMBOL_GPL(drm_of_encoder_active_endpoint);
+static int drm_of_find_remote_panel_or_bridge(struct device_node *remote,
+ struct drm_panel **panel,
+ struct drm_bridge **bridge)
+{
+ int ret = -EPROBE_DEFER;
+
+ if (panel) {
+ *panel = of_drm_find_panel(remote);
+ if (!IS_ERR(*panel))
+ ret = 0;
+ else
+ *panel = NULL;
+ }
+
+ /* 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;
+ }
+
+ }
+
+ return ret;
+}
+
/**
* drm_of_find_panel_or_bridge - return connected panel or bridge device
* @np: device tree node containing encoder output ports
@@ -249,57 +278,33 @@ int drm_of_find_panel_or_bridge(const struct device_node *np,
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;
+ /* Check for a graph on the device node first. */
+ if (of_graph_is_present(np)) {
+ remote = of_graph_get_remote_node(np, port, endpoint);
+ if (remote) {
+ ret = drm_of_find_remote_panel_or_bridge(remote, panel,
+ bridge);
+ of_node_put(remote);
+ }
}
- /*
- * 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;
+ /* Otherwise check for any child node other than port/ports. */
+ if (ret) {
+ for_each_available_child_of_node(np, remote) {
+ if (of_node_name_eq(remote, "port") ||
+ of_node_name_eq(remote, "ports"))
+ continue;
- if (panel) {
- *panel = of_drm_find_panel(remote);
- if (!IS_ERR(*panel))
- ret = 0;
- else
- *panel = NULL;
- }
+ ret = drm_of_find_remote_panel_or_bridge(remote, panel,
+ bridge);
+ of_node_put(remote);
- /* 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;
+ /* Stop at the first found occurrence. */
+ if (!ret)
+ break;
}
-
}
- of_node_put(remote);
return ret;
}
EXPORT_SYMBOL_GPL(drm_of_find_panel_or_bridge);
--
2.35.1
Hi Paul,
On Wed, Mar 09, 2022 at 03:32:00PM +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 <[email protected]>
> Fixes: 80253168dbfd ("drm: of: Lookup if child node has panel or bridge")
Thanks, it's in pretty good shape now, but I have a few bike sheds to paint :)
> ---
> drivers/gpu/drm/drm_of.c | 93 +++++++++++++++++++++-------------------
> 1 file changed, 49 insertions(+), 44 deletions(-)
>
> diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
> index 9d90cd75c457..67f1b7dfc892 100644
> --- a/drivers/gpu/drm/drm_of.c
> +++ b/drivers/gpu/drm/drm_of.c
> @@ -219,6 +219,35 @@ int drm_of_encoder_active_endpoint(struct device_node *node,
> }
> EXPORT_SYMBOL_GPL(drm_of_encoder_active_endpoint);
>
> +static int drm_of_find_remote_panel_or_bridge(struct device_node *remote,
> + struct drm_panel **panel,
> + struct drm_bridge **bridge)
This function performs its look up directly on the struct device_node
passed as argument, so I don't think the "remote" in the name is great.
Since it's static, we can just call it find_panel_or_bridge, what do you
think?
> +{
> + int ret = -EPROBE_DEFER;
> +
> + if (panel) {
> + *panel = of_drm_find_panel(remote);
> + if (!IS_ERR(*panel))
> + ret = 0;
return 0?
> + else
> + *panel = NULL;
> +
> + }
> +
> + /* No panel found yet, check for a bridge next. */
> + if (bridge) {
> + if (ret) {
And the return above allows to remove that test
> + *bridge = of_drm_find_bridge(remote);
> + if (*bridge)
> + ret = 0;
return 0?
> + } else {
> + *bridge = NULL;
> + }
> +
> + }
> +
> + return ret;
And here we can just return -EPROBE_DEFER
> +}
> +
> /**
> * drm_of_find_panel_or_bridge - return connected panel or bridge device
> * @np: device tree node containing encoder output ports
> @@ -249,57 +278,33 @@ int drm_of_find_panel_or_bridge(const struct device_node *np,
> 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;
> + /* Check for a graph on the device node first. */
> + if (of_graph_is_present(np)) {
> + remote = of_graph_get_remote_node(np, port, endpoint);
> + if (remote) {
> + ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> + bridge);
> + of_node_put(remote);
> + }
> }
>
> - /*
> - * 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;
> + /* Otherwise check for any child node other than port/ports. */
> + if (ret) {
> + for_each_available_child_of_node(np, remote) {
> + if (of_node_name_eq(remote, "port") ||
> + of_node_name_eq(remote, "ports"))
> + continue;
>
> - if (panel) {
> - *panel = of_drm_find_panel(remote);
> - if (!IS_ERR(*panel))
> - ret = 0;
> - else
> - *panel = NULL;
> - }
> + ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> + bridge);
> + of_node_put(remote);
>
> - /* 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;
> + /* Stop at the first found occurrence. */
> + if (!ret)
> + break;
> }
> -
> }
>
> - of_node_put(remote);
> return ret;
> }
So the diff is fairly hard to read, but it ends up as:
> int ret = -EPROBE_DEFER;
> struct device_node *remote;
>
> if (!panel && !bridge)
> return -EINVAL;
> if (panel)
> *panel = NULL;
>
> /* Check for a graph on the device node first. */
> if (of_graph_is_present(np)) {
> remote = of_graph_get_remote_node(np, port, endpoint);
> if (remote) {
> ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> bridge);
> of_node_put(remote);
I think we can simplify this by doing
if (!ret)
return ret;
> }
> }
>
> /* Otherwise check for any child node other than port/ports. */
> if (ret) {
And thus we won't have to check for ret here
> for_each_available_child_of_node(np, remote) {
I'm a bit reluctant with variables that we reuse from one loop to
another, especially since it's a bit misleading here. What about using a
(loop local) remote variable in the of_graph path, and a loop-local
variable node or child here?
> if (of_node_name_eq(remote, "port") ||
> of_node_name_eq(remote, "ports"))
> continue;
>
> ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> bridge);
> of_node_put(remote);
>
> /* Stop at the first found occurrence. */
> if (!ret)
> break;
Ditto, let's just return here
> }
> }
>
> return ret;
And then we can just return EPROBE_DEFER here (and get rid of ret entirely)
Maxime
Hi Maxime,
Thanks for the review!
On Thu 10 Mar 22, 15:54, Maxime Ripard wrote:
> Hi Paul,
>
> On Wed, Mar 09, 2022 at 03:32:00PM +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 <[email protected]>
> > Fixes: 80253168dbfd ("drm: of: Lookup if child node has panel or bridge")
>
> Thanks, it's in pretty good shape now, but I have a few bike sheds to paint :)
>
> > ---
> > drivers/gpu/drm/drm_of.c | 93 +++++++++++++++++++++-------------------
> > 1 file changed, 49 insertions(+), 44 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
> > index 9d90cd75c457..67f1b7dfc892 100644
> > --- a/drivers/gpu/drm/drm_of.c
> > +++ b/drivers/gpu/drm/drm_of.c
> > @@ -219,6 +219,35 @@ int drm_of_encoder_active_endpoint(struct device_node *node,
> > }
> > EXPORT_SYMBOL_GPL(drm_of_encoder_active_endpoint);
> >
> > +static int drm_of_find_remote_panel_or_bridge(struct device_node *remote,
> > + struct drm_panel **panel,
> > + struct drm_bridge **bridge)
>
> This function performs its look up directly on the struct device_node
> passed as argument, so I don't think the "remote" in the name is great.
> Since it's static, we can just call it find_panel_or_bridge, what do you
> think?
From a quick look at other DRM code I got the impression that static functions
also usually carry the drm prefix but I might be wrong.
> > +{
> > + int ret = -EPROBE_DEFER;
> > +
> > + if (panel) {
> > + *panel = of_drm_find_panel(remote);
> > + if (!IS_ERR(*panel))
> > + ret = 0;
>
> return 0?
The idea was to still go through the "*bridge = NULL;" path if a bridge
pointer is provided, to preserve the original behavior of the function.
There may or may not not be any hard expectation on that, in any case
I feel like it would be good to avoid out-of-scope functional changes here.
> > + else
> > + *panel = NULL;
> > +
> > + }
> > +
> > + /* No panel found yet, check for a bridge next. */
> > + if (bridge) {
> > + if (ret) {
>
> And the return above allows to remove that test
>
> > + *bridge = of_drm_find_bridge(remote);
> > + if (*bridge)
> > + ret = 0;
>
> return 0?
>
> > + } else {
> > + *bridge = NULL;
> > + }
> > +
> > + }
> > +
> > + return ret;
>
> And here we can just return -EPROBE_DEFER
>
> > +}
> > +
>
> > /**
> > * drm_of_find_panel_or_bridge - return connected panel or bridge device
> > * @np: device tree node containing encoder output ports
> > @@ -249,57 +278,33 @@ int drm_of_find_panel_or_bridge(const struct device_node *np,
> > 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;
> > + /* Check for a graph on the device node first. */
> > + if (of_graph_is_present(np)) {
> > + remote = of_graph_get_remote_node(np, port, endpoint);
> > + if (remote) {
> > + ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> > + bridge);
> > + of_node_put(remote);
> > + }
> > }
> >
> > - /*
> > - * 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;
> > + /* Otherwise check for any child node other than port/ports. */
> > + if (ret) {
> > + for_each_available_child_of_node(np, remote) {
> > + if (of_node_name_eq(remote, "port") ||
> > + of_node_name_eq(remote, "ports"))
> > + continue;
> >
> > - if (panel) {
> > - *panel = of_drm_find_panel(remote);
> > - if (!IS_ERR(*panel))
> > - ret = 0;
> > - else
> > - *panel = NULL;
> > - }
> > + ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> > + bridge);
> > + of_node_put(remote);
> >
> > - /* 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;
> > + /* Stop at the first found occurrence. */
> > + if (!ret)
> > + break;
> > }
> > -
> > }
> >
> > - of_node_put(remote);
> > return ret;
> > }
>
> So the diff is fairly hard to read, but it ends up as:
Yeah I agree, not sure what I can do about that.
> > int ret = -EPROBE_DEFER;
> > struct device_node *remote;
> >
> > if (!panel && !bridge)
> > return -EINVAL;
> > if (panel)
> > *panel = NULL;
> >
> > /* Check for a graph on the device node first. */
> > if (of_graph_is_present(np)) {
> > remote = of_graph_get_remote_node(np, port, endpoint);
> > if (remote) {
> > ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> > bridge);
> > of_node_put(remote);
>
> I think we can simplify this by doing
>
> if (!ret)
> return ret;
>
> > }
> > }
> >
> > /* Otherwise check for any child node other than port/ports. */
> > if (ret) {
>
> And thus we won't have to check for ret here
Yes I agree this one makes things more readable.
> > for_each_available_child_of_node(np, remote) {
>
> I'm a bit reluctant with variables that we reuse from one loop to
> another, especially since it's a bit misleading here. What about using a
> (loop local) remote variable in the of_graph path, and a loop-local
> variable node or child here?
I feel like reusing variables across loops is quite a common thing and
not really an issue on its own, but I agree that calling this one remote
is confusing and "child" would make things clearer here.
> > if (of_node_name_eq(remote, "port") ||
> > of_node_name_eq(remote, "ports"))
> > continue;
> >
> > ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> > bridge);
> > of_node_put(remote);
> >
> > /* Stop at the first found occurrence. */
> > if (!ret)
> > break;
>
> Ditto, let's just return here
Sure, fair enough!
> > }
> > }
> >
> > return ret;
>
> And then we can just return EPROBE_DEFER here (and get rid of ret entirely)
Sounds good to me, thanks!
Paul
--
Paul Kocialkowski, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com
Hi Jagan,
On Fri 18 Mar 22, 21:35, Jagan Teki wrote:
> Hi Paul,
>
> On Wed, Mar 9, 2022 at 8:02 PM Paul Kocialkowski
> <[email protected]> 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.
>
> Can you add that pipeline example on the commit message, it gives more
> information on specific use cases why the existing code breaks.
Ah I just sent v2 before reading your message.
Well I think the description says it all: the problem shows as soon as there's
a child node to the node passed to drm_of_find_panel_or_bridge and it's really
independent from the of graph setup in the end.
I think Maxime put some examples on the original thread (v4 of your patch).
Paul
--
Paul Kocialkowski, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com
On Wed, Mar 16, 2022 at 04:40:49PM +0100, Paul Kocialkowski wrote:
> Hi Maxime,
>
> Thanks for the review!
>
> On Thu 10 Mar 22, 15:54, Maxime Ripard wrote:
> > Hi Paul,
> >
> > On Wed, Mar 09, 2022 at 03:32:00PM +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 <[email protected]>
> > > Fixes: 80253168dbfd ("drm: of: Lookup if child node has panel or bridge")
> >
> > Thanks, it's in pretty good shape now, but I have a few bike sheds to paint :)
> >
> > > ---
> > > drivers/gpu/drm/drm_of.c | 93 +++++++++++++++++++++-------------------
> > > 1 file changed, 49 insertions(+), 44 deletions(-)
> > >
> > > diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
> > > index 9d90cd75c457..67f1b7dfc892 100644
> > > --- a/drivers/gpu/drm/drm_of.c
> > > +++ b/drivers/gpu/drm/drm_of.c
> > > @@ -219,6 +219,35 @@ int drm_of_encoder_active_endpoint(struct device_node *node,
> > > }
> > > EXPORT_SYMBOL_GPL(drm_of_encoder_active_endpoint);
> > >
> > > +static int drm_of_find_remote_panel_or_bridge(struct device_node *remote,
> > > + struct drm_panel **panel,
> > > + struct drm_bridge **bridge)
> >
> > This function performs its look up directly on the struct device_node
> > passed as argument, so I don't think the "remote" in the name is great.
> > Since it's static, we can just call it find_panel_or_bridge, what do you
> > think?
>
> From a quick look at other DRM code I got the impression that static functions
> also usually carry the drm prefix but I might be wrong.
Not necessarily, see handle_conflicting_encoders, commit_tail, commit_work,
convert_clip_rect_to_rect, edid_load, etc.
Most functions do, but it's not a rule or a convention.
> > > +{
> > > + int ret = -EPROBE_DEFER;
> > > +
> > > + if (panel) {
> > > + *panel = of_drm_find_panel(remote);
> > > + if (!IS_ERR(*panel))
> > > + ret = 0;
> >
> > return 0?
>
> The idea was to still go through the "*bridge = NULL;" path if a bridge
> pointer is provided, to preserve the original behavior of the function.
> There may or may not not be any hard expectation on that, in any case
> I feel like it would be good to avoid out-of-scope functional changes here.
Then we could just clear it just like we clear the panel pointer in
drm_of_find_panel_or_bridge. It would be more consistent.
> > > + else
> > > + *panel = NULL;
> > > +
> > > + }
> > > +
> > > + /* No panel found yet, check for a bridge next. */
> > > + if (bridge) {
> > > + if (ret) {
> >
> > And the return above allows to remove that test
> >
> > > + *bridge = of_drm_find_bridge(remote);
> > > + if (*bridge)
> > > + ret = 0;
> >
> > return 0?
> >
> > > + } else {
> > > + *bridge = NULL;
> > > + }
> > > +
> > > + }
> > > +
> > > + return ret;
> >
> > And here we can just return -EPROBE_DEFER
> >
> > > +}
> > > +
> >
> > > /**
> > > * drm_of_find_panel_or_bridge - return connected panel or bridge device
> > > * @np: device tree node containing encoder output ports
> > > @@ -249,57 +278,33 @@ int drm_of_find_panel_or_bridge(const struct device_node *np,
> > > 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;
> > > + /* Check for a graph on the device node first. */
> > > + if (of_graph_is_present(np)) {
> > > + remote = of_graph_get_remote_node(np, port, endpoint);
> > > + if (remote) {
> > > + ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> > > + bridge);
> > > + of_node_put(remote);
> > > + }
> > > }
> > >
> > > - /*
> > > - * 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;
> > > + /* Otherwise check for any child node other than port/ports. */
> > > + if (ret) {
> > > + for_each_available_child_of_node(np, remote) {
> > > + if (of_node_name_eq(remote, "port") ||
> > > + of_node_name_eq(remote, "ports"))
> > > + continue;
> > >
> > > - if (panel) {
> > > - *panel = of_drm_find_panel(remote);
> > > - if (!IS_ERR(*panel))
> > > - ret = 0;
> > > - else
> > > - *panel = NULL;
> > > - }
> > > + ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> > > + bridge);
> > > + of_node_put(remote);
> > >
> > > - /* 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;
> > > + /* Stop at the first found occurrence. */
> > > + if (!ret)
> > > + break;
> > > }
> > > -
> > > }
> > >
> > > - of_node_put(remote);
> > > return ret;
> > > }
> >
> > So the diff is fairly hard to read, but it ends up as:
>
> Yeah I agree, not sure what I can do about that.
Nothing, really. I don't expect any change there, it just happens sometimes :)
Maxime
Hi Maxime,
On Fri 18 Mar 22, 16:14, Maxime Ripard wrote:
> On Wed, Mar 16, 2022 at 04:40:49PM +0100, Paul Kocialkowski wrote:
> > Hi Maxime,
> >
> > Thanks for the review!
> >
> > On Thu 10 Mar 22, 15:54, Maxime Ripard wrote:
> > > Hi Paul,
> > >
> > > On Wed, Mar 09, 2022 at 03:32:00PM +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 <[email protected]>
> > > > Fixes: 80253168dbfd ("drm: of: Lookup if child node has panel or bridge")
> > >
> > > Thanks, it's in pretty good shape now, but I have a few bike sheds to paint :)
> > >
> > > > ---
> > > > drivers/gpu/drm/drm_of.c | 93 +++++++++++++++++++++-------------------
> > > > 1 file changed, 49 insertions(+), 44 deletions(-)
> > > >
> > > > diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
> > > > index 9d90cd75c457..67f1b7dfc892 100644
> > > > --- a/drivers/gpu/drm/drm_of.c
> > > > +++ b/drivers/gpu/drm/drm_of.c
> > > > @@ -219,6 +219,35 @@ int drm_of_encoder_active_endpoint(struct device_node *node,
> > > > }
> > > > EXPORT_SYMBOL_GPL(drm_of_encoder_active_endpoint);
> > > >
> > > > +static int drm_of_find_remote_panel_or_bridge(struct device_node *remote,
> > > > + struct drm_panel **panel,
> > > > + struct drm_bridge **bridge)
> > >
> > > This function performs its look up directly on the struct device_node
> > > passed as argument, so I don't think the "remote" in the name is great.
> > > Since it's static, we can just call it find_panel_or_bridge, what do you
> > > think?
> >
> > From a quick look at other DRM code I got the impression that static functions
> > also usually carry the drm prefix but I might be wrong.
>
> Not necessarily, see handle_conflicting_encoders, commit_tail, commit_work,
> convert_clip_rect_to_rect, edid_load, etc.
>
> Most functions do, but it's not a rule or a convention.
Okay then, I'm fine with find_panel_or_bridge.
> > > > +{
> > > > + int ret = -EPROBE_DEFER;
> > > > +
> > > > + if (panel) {
> > > > + *panel = of_drm_find_panel(remote);
> > > > + if (!IS_ERR(*panel))
> > > > + ret = 0;
> > >
> > > return 0?
> >
> > The idea was to still go through the "*bridge = NULL;" path if a bridge
> > pointer is provided, to preserve the original behavior of the function.
> > There may or may not not be any hard expectation on that, in any case
> > I feel like it would be good to avoid out-of-scope functional changes here.
>
> Then we could just clear it just like we clear the panel pointer in
> drm_of_find_panel_or_bridge. It would be more consistent.
Oh absolutely, I agree that would be best.
> > > > + else
> > > > + *panel = NULL;
> > > > +
> > > > + }
> > > > +
> > > > + /* No panel found yet, check for a bridge next. */
> > > > + if (bridge) {
> > > > + if (ret) {
> > >
> > > And the return above allows to remove that test
> > >
> > > > + *bridge = of_drm_find_bridge(remote);
> > > > + if (*bridge)
> > > > + ret = 0;
> > >
> > > return 0?
> > >
> > > > + } else {
> > > > + *bridge = NULL;
> > > > + }
> > > > +
> > > > + }
> > > > +
> > > > + return ret;
> > >
> > > And here we can just return -EPROBE_DEFER
> > >
> > > > +}
> > > > +
> > >
> > > > /**
> > > > * drm_of_find_panel_or_bridge - return connected panel or bridge device
> > > > * @np: device tree node containing encoder output ports
> > > > @@ -249,57 +278,33 @@ int drm_of_find_panel_or_bridge(const struct device_node *np,
> > > > 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;
> > > > + /* Check for a graph on the device node first. */
> > > > + if (of_graph_is_present(np)) {
> > > > + remote = of_graph_get_remote_node(np, port, endpoint);
> > > > + if (remote) {
> > > > + ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> > > > + bridge);
> > > > + of_node_put(remote);
> > > > + }
> > > > }
> > > >
> > > > - /*
> > > > - * 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;
> > > > + /* Otherwise check for any child node other than port/ports. */
> > > > + if (ret) {
> > > > + for_each_available_child_of_node(np, remote) {
> > > > + if (of_node_name_eq(remote, "port") ||
> > > > + of_node_name_eq(remote, "ports"))
> > > > + continue;
> > > >
> > > > - if (panel) {
> > > > - *panel = of_drm_find_panel(remote);
> > > > - if (!IS_ERR(*panel))
> > > > - ret = 0;
> > > > - else
> > > > - *panel = NULL;
> > > > - }
> > > > + ret = drm_of_find_remote_panel_or_bridge(remote, panel,
> > > > + bridge);
> > > > + of_node_put(remote);
> > > >
> > > > - /* 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;
> > > > + /* Stop at the first found occurrence. */
> > > > + if (!ret)
> > > > + break;
> > > > }
> > > > -
> > > > }
> > > >
> > > > - of_node_put(remote);
> > > > return ret;
> > > > }
> > >
> > > So the diff is fairly hard to read, but it ends up as:
> >
> > Yeah I agree, not sure what I can do about that.
>
> Nothing, really. I don't expect any change there, it just happens sometimes :)
All right then :)
I'll send another iteration soon.
Cheers,
Paul
--
Paul Kocialkowski, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com
Hi Paul,
On Wed, Mar 9, 2022 at 8:02 PM Paul Kocialkowski
<[email protected]> 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.
Can you add that pipeline example on the commit message, it gives more
information on specific use cases why the existing code breaks.
Thanks,
Jagan.