On 2018-04-19 18:22, Rob Herring wrote: > On Tue, Apr 17, 2018 at 8:10 AM, Peter Rosin <peda@xxxxxxxxxx> wrote: >> Add a central function to parse a node according to the video >> interface binding and get a media bus format. >> >> Start with only supporting a very limited set of a few basic media >> bus formats. >> >> Signed-off-by: Peter Rosin <peda@xxxxxxxxxx> >> --- >> drivers/gpu/drm/drm_of.c | 38 ++++++++++++++++++++++++++++++++++++++ >> include/drm/drm_of.h | 7 +++++++ >> 2 files changed, 45 insertions(+) >> >> diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c >> index 4c191c050e7d..f9473edb60a7 100644 >> --- a/drivers/gpu/drm/drm_of.c >> +++ b/drivers/gpu/drm/drm_of.c >> @@ -212,6 +212,44 @@ int drm_of_encoder_active_endpoint(struct device_node *node, >> EXPORT_SYMBOL_GPL(drm_of_encoder_active_endpoint); >> >> /* >> + * drm_of_media_bus_fmt - return the media bus format described in the node >> + * @node: device tree node containing the media bus format >> + * >> + * @node is presumably an of-graph endpoint node. >> + * >> + * Return the media bus format, or zero if none is described. Or one of the >> + * standard error codes. >> + */ >> +int drm_of_media_bus_fmt(struct device_node *node) > > Should use endpoint or ep here instead of node to be clear what node > this function operates on. I just sent v3 and missed this. Anyway, the reason I didn't use ep, was that you at one point said "port of endpoint", and I wanted this function to cover both cases. And maybe others. Anyway, if I hear nothing I'll make the change to ep and I'll also reword the comment to talk more decisively about endpoints for v4. Cheers, Peter _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/dri-devel
