v4l2_pipeline_link_notify(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

okay so bestie, we're in the Kernel API Manual today and this one's a certified throwback β€” it's giving "senior dev who's seen things" energy. Buckle up πŸŽ€


✨ NAME ✨

v4l2_pipeline_link_notify β€” the link management notification callback πŸ¦‹


🌸 SYNOPSIS 🌸

int v4l2_pipeline_link_notify (struct media_link *link , u32 flags ,
       unsigned int notification );

πŸ’– ARGUMENTS πŸ’–

  • link β€” the link, bestie. that's the whole description. it's THE link πŸ’…
  • flags β€” the new link flags that are gonna get applied ✨
  • notification β€” the link's state change notification type, i.e. one of the MEDIA_DEV_NOTIFY_* fam 🫢

πŸ”₯ DESCRIPTION πŸ”₯

THIS FUNCTION IS DEPRECATED. DO NOT USE IN NEW DRIVERS. USE RUNTIME PM ON SUB-DEVICE DRIVERS INSTEAD. πŸ’€

like. they yelled it in all caps for a reason, ok? not me being the messenger πŸ™ˆ this one is legacy. new drivers, do not touch. use runtime PM on the sub-device drivers instead, periodt.

so what does it even do? it reacts to link management happening on powered pipelines by updating the use count of every entity on the source side and the sink side of the link. entities get powered on or off accordingly β€” main character energy for the whole graph πŸ‘‘

and here's the pairing rule: the use of this function should be paired with v4l2_pipeline_pm_{get,put}(). they're a set. don't split them up πŸŽ€

it returns 0 on success, or a negative error code on failure. and ngl, the little optimism moment here: powering entities off is assumed to never fail ✨. also, this function will not fail for disconnection events β€” she simply does not care πŸ’…


🌈 SEE ALSO 🌈

a whole friend group, fr fr 🫢


and that's the page, bestie! πŸ¦‹ this iconic little artifact lives in ./include/media and got its timestamp of October 2026 β€” shoutout to the legends who documented it, mwah πŸ’‹ <rainbow>v4l2_pipeline_link_notify(9) out πŸ’–</rainbow>

πŸ“¦ where this page comes from