On Wed, Nov 18, 2020 at 4:09 PM Simon Ser contact@emersion.fr wrote:
On Wednesday, November 18, 2020 4:03 PM, Daniel Vetter daniel@ffwll.ch wrote:
I think this causes warnings, because now we have kerneldoc for this, but not for all the members. Also the member-specific stuff should be documented as inline comment, see
https://dri.freedesktop.org/docs/drm/doc-guide/kernel-doc.html#in-line-membe...
Hm, right, will make sure the patch doesn't trigger warnings.
I still think the force-probe stuff shouldn't be documented in in-line comments, because I'd never look at the in-line count_modes comment to know whether the ioctl probes or not. Adding short in-line comments sounds fine though.
Hm yeah, maybe give it a sub-title then like "Probe Modes" with the two options. I think the detail that a small array/count (either 0 or 1 depending whether you want force probing) should be documented in the inline comment though.
I think the inline comment for connection should also link to the drm_connector_status enum. Maybe also have that link in the "Probe Modes" section with an explanation that a forced probe can get rid of some the unknown ones.
I also kinda wonder where we should up recommendations for autoconfiguration.
I also noticed that this file has a ton of wrong kerneldoc comments, but they seem to not cause warnings (anything starting with /** is fishy).
Can I volunteer you for a bit more here?
Yeah, I've noticed this as well. Will have a look!
Thanks! -Daniel