Hi Johannes, Thank you for the patch. On Tue, May 23, 2023 at 10:19:43AM +0200, Johannes Zink wrote: > Some panels support multiple LVDS data mapping formats, which can be > used e.g. run displays on jeida-18 format when only 3 LVDS lanes are > available. > > Add parsing of an optional data-mapping devicetree property, which also > touches up the bits per color to match the bus format. Of course one could argue that the innolux,g101ice-l01 panel should have used the panel-lvds bindings... :-) > Signed-off-by: Johannes Zink <j.zink@xxxxxxxxxxxxxx> > > --- > > Changes: > > v1 -> v2: - fix missing unwind goto found by test robot > Reported-by: kernel test robot <lkp@xxxxxxxxx> > Reported-by: Dan Carpenter <error27@xxxxxxxxx> > Link: https://lore.kernel.org/r/202304160359.4LHmFOlU-lkp@xxxxxxxxx/ > --- > drivers/gpu/drm/panel/panel-simple.c | 39 +++++++++++++++++++++++++++++++++++- > 1 file changed, 38 insertions(+), 1 deletion(-) > > diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c > index 2a9c1a785a5c..0a35fdb49ccb 100644 > --- a/drivers/gpu/drm/panel/panel-simple.c > +++ b/drivers/gpu/drm/panel/panel-simple.c > @@ -40,6 +40,7 @@ > #include <drm/drm_edid.h> > #include <drm/drm_mipi_dsi.h> > #include <drm/drm_panel.h> > +#include <drm/drm_of.h> > > /** > * struct panel_desc - Describes a simple panel. > @@ -559,7 +560,7 @@ static int panel_simple_probe(struct device *dev, const struct panel_desc *desc) > struct device_node *ddc; > int connector_type; > u32 bus_flags; > - int err; > + int err, ret; > > panel = devm_kzalloc(dev, sizeof(*panel), GFP_KERNEL); > if (!panel) > @@ -605,6 +606,42 @@ static int panel_simple_probe(struct device *dev, const struct panel_desc *desc) > panel_simple_parse_panel_timing_node(dev, panel, &dt); > } > > + Double blank line. > + /* optional data-mapping property for overriding bus format */ s/optional/Optional/ > + ret = drm_of_lvds_get_data_mapping(dev->of_node); > + if (ret == -EINVAL) { > + dev_warn(dev, "Ignore invalid data-mapping property"); > + } else if (ret != -ENODEV) { If someone incorrectly sets the property in DT for a non-LVDS panel, the result won't be nice. That's of course a DT issue, but I wonder if we could/should protect against it. You could move this code to a separate function (which would have the added benefit of lowering the indentation level as you can return early in error cases), and call it from panel_simple_probe() only if the panel is an LVDS panel (as reported by its desc->bus_format value). > + int bpc; > + > + switch (ret) { > + default: > + WARN_ON(1); > + fallthrough; > + case MEDIA_BUS_FMT_RGB888_1X7X4_SPWG: > + fallthrough; > + case MEDIA_BUS_FMT_RGB888_1X7X4_JEIDA: > + bpc = 8; > + break; > + case MEDIA_BUS_FMT_RGB666_1X7X3_SPWG: > + bpc = 6; > + } > + > + if (desc->bpc != bpc || desc->bus_format != ret) { > + struct panel_desc *override_desc; > + > + override_desc = devm_kmemdup(dev, desc, sizeof(*desc), GFP_KERNEL); > + if (!override_desc) { > + err = -ENOMEM; > + goto free_ddc; > + } > + > + override_desc->bus_format = ret; > + override_desc->bpc = bpc; > + panel->desc = override_desc; > + } > + } > + > connector_type = desc->connector_type; > /* Catch common mistakes for panels. */ > switch (connector_type) { -- Regards, Laurent Pinchart
