[PATCH v9 1/4] drm: of: Add drm_of_get_dsi_bus helper function

[Chris Morgan]

On Wed, Jan 11, 2023 at 05:39:26PM +0100, Maxime Ripard wrote:
> Hi,
> 
> On Wed, Jan 11, 2023 at 10:30:09AM -0600, Chris Morgan wrote:
> > From: Chris Morgan <macromorgan at hotmail.com>
> > 
> > Add helper function to find DSI host for devices where DSI panel is not
> > a minor of a DSI bus (such as the Samsung AMS495QA01 panel or the
> > official Raspberry Pi touchscreen display).
> > 
> > Signed-off-by: Chris Morgan <macromorgan at hotmail.com>
> > Signed-off-by: Maya Matuszczyk <maccraft123mc at gmail.com>
> > ---
> >  drivers/gpu/drm/drm_of.c | 70 ++++++++++++++++++++++++++++++++++++++++
> >  include/drm/drm_of.h     | 10 ++++++
> >  2 files changed, 80 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
> > index 7bbcb999bb75..4ebb5bc4b595 100644
> > --- a/drivers/gpu/drm/drm_of.c
> > +++ b/drivers/gpu/drm/drm_of.c
> > @@ -10,6 +10,7 @@
> >  #include <drm/drm_crtc.h>
> >  #include <drm/drm_device.h>
> >  #include <drm/drm_encoder.h>
> > +#include <drm/drm_mipi_dsi.h>
> >  #include <drm/drm_of.h>
> >  #include <drm/drm_panel.h>
> >  
> > @@ -493,3 +494,72 @@ int drm_of_get_data_lanes_count_ep(const struct device_node *port,
> >  	return ret;
> >  }
> >  EXPORT_SYMBOL_GPL(drm_of_get_data_lanes_count_ep);
> > +
> > +/**
> > + * drm_of_get_dsi_bus - find the DSI bus for a given device
> > + * @dev: parent device of display (SPI, I2C)
> > + * @info: DSI device info to be updated with correct DSI node
> > + *
> > + * Gets parent DSI bus for a DSI device controlled through a bus other
> > + * than MIPI-DCS (SPI, I2C, etc.) using the Device Tree.
> > + *
> > + * Returns pointer to mipi_dsi_host if successful, -EINVAL if the
> > + * request is unsupported, -EPROBE_DEFER if the DSI host is found but
> > + * not available, or -ENODEV otherwise.
> > + */
> > +struct mipi_dsi_host *drm_of_get_dsi_bus(struct device *dev,
> > +					 struct mipi_dsi_device_info *info)
> > +{
> > +	struct mipi_dsi_host *dsi_host;
> > +	struct device_node *endpoint, *dsi_host_node;
> > +
> > +	/*
> > +	 * Exit immediately if we attempt to call this function when
> > +	 * DRM_MIPI_DSI is not enabled, in the event CONFIG_OF is
> > +	 * enabled.
> > +	 */
> > +	if (!IS_ENABLED(CONFIG_DRM_MIPI_DSI))
> > +		return ERR_PTR(-EINVAL);
> 
> The commit log isn't super clear on why this is needed, but it would be
> more consistent to add an ifdef and only compile the entire function if
> DRM_MIPI_DSI is there, just like you did for OF already.

Thank you. I can do that, I just wasn't sure if "#ifdefs" were frowned
upon or not. That would probably be the easiest way to do this though.

> 
> > +	/*
> > +	 * Get first endpoint child from device.
> > +	 */
> > +	endpoint = of_graph_get_next_endpoint(dev->of_node, NULL);
> > +	if (!endpoint)
> > +		return ERR_PTR(-ENODEV);
> > +
> > +	/*
> > +	 * Follow the first endpoint to get the DSI host node.
> > +	 */
> > +	dsi_host_node = of_graph_get_remote_port_parent(endpoint);
> 
> There's no need to hold the reference to endpoint after that call. call
> of_node_put(endpoint) here, and it will simplify the error path.
> 
> > +	if (!dsi_host_node)
> > +		goto error;
> > +
> > +	/*
> > +	 * Get the DSI host from the DSI host node. If we get an error
> > +	 * or the return is null assume we're not ready to probe just
> > +	 * yet. Release the DSI host node since we're done with it.
> > +	 */
> > +	dsi_host = of_find_mipi_dsi_host_by_node(dsi_host_node);
> > +	of_node_put(dsi_host_node);
> > +	if (IS_ERR_OR_NULL(dsi_host)) {
> > +		of_node_put(endpoint);
> > +		return ERR_PTR(-EPROBE_DEFER);
> > +	}
> > +
> > +	/*
> > +	 * Set the node of the mipi_dsi_device_info to the correct node
> > +	 * and then release the endpoint node since we're done with it.
> > +	 */
> > +	info->node = of_graph_get_remote_port(endpoint);
> 
> Ah, you're using it there.
> 
> I think I'd rework the function to:
> 
> - retrieve the endpoint
> - retrieve the remote port, give up the endpoint
> - retrieve the remote port parent
> 
> Also, I'm not entirely sure what you had in mind, but info might not be
> there at all and it would be fine imho.
> 

What if I make it optional and if a NULL is passed skip this step, but
otherwise populate the info node?

> Maxime

Thank you for your input. I'll wait for the bots to see if the previous
errors are really finally fixed, then I'll make these changes and
resubmit.