Re: [PATCH v3 5/6] drm: lcdif: Add multiple encoders and first bridges support

[Liu Ying <victor.liu@nxp.com>]

On Wed, 2023-02-15 at 08:54 +0100, Alexander Stein wrote:
> Hi Liu,

Hi Alexander,

> 
> thanks for the update.

Thanks for the review.

> 
> Am Montag, 13. Februar 2023, 09:56:11 CET schrieb Liu Ying:
> > The single LCDIF embedded in i.MX93 SoC may drive multiple displays
> > simultaneously.  Look at LCDIF output port's remote port parents to
> > find all enabled first bridges.  Add an encoder for each found
> > bridge
> > and attach the bridge to the encoder.  This is a preparation for
> > adding i.MX93 LCDIF support.
> > 
> > Signed-off-by: Liu Ying <victor.liu@nxp.com>
> > ---
> > v2->v3:
> > * No change.
> > 
> > v1->v2:
> > * Split from patch 2/2 in v1. (Marek, Alexander)
> > * Drop '!remote ||' from lcdif_attach_bridge(). (Lothar)
> > * Drop unneeded 'bridges' member from lcdif_drm_private structure.
> > 
> >  drivers/gpu/drm/mxsfb/lcdif_drv.c | 68
> > +++++++++++++++++++++++++++----
> >  drivers/gpu/drm/mxsfb/lcdif_drv.h |  4 +-
> >  drivers/gpu/drm/mxsfb/lcdif_kms.c | 21 ++--------
> >  3 files changed, 66 insertions(+), 27 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/mxsfb/lcdif_drv.c
> > b/drivers/gpu/drm/mxsfb/lcdif_drv.c index
> > b5b9a8e273c6..eb6c265fa2fe 100644
> > --- a/drivers/gpu/drm/mxsfb/lcdif_drv.c
> > +++ b/drivers/gpu/drm/mxsfb/lcdif_drv.c
> > @@ -9,13 +9,16 @@
> >  #include <linux/dma-mapping.h>
> >  #include <linux/io.h>
> >  #include <linux/module.h>
> > +#include <linux/of.h>
> >  #include <linux/of_device.h>
> > +#include <linux/of_graph.h>
> >  #include <linux/platform_device.h>
> >  #include <linux/pm_runtime.h>
> > 
> >  #include <drm/drm_atomic_helper.h>
> >  #include <drm/drm_bridge.h>
> >  #include <drm/drm_drv.h>
> > +#include <drm/drm_encoder.h>
> >  #include <drm/drm_fbdev_generic.h>
> >  #include <drm/drm_gem_dma_helper.h>
> >  #include <drm/drm_gem_framebuffer_helper.h>
> > @@ -38,19 +41,68 @@ static const struct
> > drm_mode_config_helper_funcs
> > lcdif_mode_config_helpers = { .atomic_commit_tail =
> > drm_atomic_helper_commit_tail_rpm,
> >  };
> > 
> > +static const struct drm_encoder_funcs lcdif_encoder_funcs = {
> > +	.destroy = drm_encoder_cleanup,
> > +};
> > +
> >  static int lcdif_attach_bridge(struct lcdif_drm_private *lcdif)
> >  {
> > -	struct drm_device *drm = lcdif->drm;
> > +	struct device *dev = lcdif->drm->dev;
> > +	struct device_node *ep;
> >  	struct drm_bridge *bridge;
> >  	int ret;
> > 
> > -	bridge = devm_drm_of_get_bridge(drm->dev, drm->dev->of_node, 0,
> > 0);
> > -	if (IS_ERR(bridge))
> > -		return PTR_ERR(bridge);
> > -
> > -	ret = drm_bridge_attach(&lcdif->encoder, bridge, NULL, 0);
> > -	if (ret)
> > -		return dev_err_probe(drm->dev, ret, "Failed to attach 
> 
> bridge\n");
> > +	for_each_endpoint_of_node(dev->of_node, ep) {
> > +		struct device_node *remote;
> > +		struct of_endpoint of_ep;
> > +		struct drm_encoder *encoder;
> > +
> > +		remote = of_graph_get_remote_port_parent(ep);
> 
> Is it possible for remote to be NULL?

Yes. But, no worries.  Lothar said the check for '!remote' is not
needed in v1 comment because of_device_is_available() checks that.

> 
> > +		if (!of_device_is_available(remote)) {
> > +			of_node_put(remote);
> > +			continue;
> > +		}
> > +		of_node_put(remote);
> > +
> > +		ret = of_graph_parse_endpoint(ep, &of_ep);
> > +		if (ret < 0) {
> > +			dev_err(dev, "Failed to parse endpoint
> > %pOF\n", 
> 
> ep);
> > +			of_node_put(ep);
> > +			return ret;
> > +		}
> > +
> > +		if (of_ep.id >= MAX_DISPLAYS) {
> > +			dev_warn(dev, "invalid endpoint id %u\n", 
> 
> of_ep.id);
> 
> I would write
> dev_warn(dev, "ignoring invalid endpoint id %u\n", of_ep.id);
> just because the parsing continues but this one is skipped.

Ok, will do that in next version.

> 
> > +			continue;
> > +		}
> > +
> > +		bridge = devm_drm_of_get_bridge(dev, dev->of_node, 0, 
> 
> of_ep.id);
> > +		if (IS_ERR(bridge)) {
> > +			of_node_put(ep);
> > +			return dev_err_probe(dev, PTR_ERR(bridge),
> > +					     "Failed to get bridge 
> 
> for endpoint%u\n",
> > +					     of_ep.id);
> > +		}
> > +
> > +		encoder = &lcdif->encoders[of_ep.id];
> > +		encoder->possible_crtcs = drm_crtc_mask(&lcdif->crtc);
> > +		ret = drm_encoder_init(lcdif->drm, encoder, 
> 
> &lcdif_encoder_funcs,
> > +				       DRM_MODE_ENCODER_NONE, NULL);
> > +		if (ret) {
> > +			dev_err(dev, "Failed to initialize encoder for 
> 
> endpoint%u: %d\n",
> > +				of_ep.id, ret);
> > +			of_node_put(ep);
> > +			return ret;
> > +		}
> > +
> > +		ret = drm_bridge_attach(encoder, bridge, NULL, 0);
> > +		if (ret) {
> > +			of_node_put(ep);
> > +			return dev_err_probe(dev, ret,
> > +					     "Failed to attach 
> 
> bridge for endpoint%u\n",
> > +					     of_ep.id);
> > +		}
> 
> Admittedly I'm not used to the drm API, but do we need to some manual
> cleanup/
> revert if some endpoints is e.g. deferred, but previous endpoints
> already have 
> been successfully added? e.g. endpoint 0 is added, but adding
> endpoint 1 
> fails.

I think the bailout path is safe, because
drm_mode_config_init_release() is called as the managed release action
added through drm_mode_config_init(). drm_mode_config_cleanup() would
clean things up.

Regards,
Liu Ying