__devlink_re...pshot_id_get(9)

📦 version 7.3.1, in nixpkgs unstable

okay bestie so welcome to the Kernel API Manual 💖 you're literally reading the page for __devlink_re...pshot_id_get(9) and it's giving kernel-internals realness, so let's get into it 🎀

✨ NAME ✨

__devlink_region_snapshot_id_get — get snapshot ID 🫶 (that's the whole job, she's a getter, an icon)

✨ SYNOPSIS ✨

int __devlink_region_snapshot_id_get (struct devlink *devlink , u32 *id );

✨ ARGUMENTS ✨

  • devlink — the devlink instance, like the actual object you're working with, no cap 💅
  • id — storage to return snapshot id — that's the lil spot where the snapshot id gets handed back to you 🎀

✨ DESCRIPTION ✨

okay so what she does is allocates a new snapshot id ✨ Returns zero on success, or a negative error on failure 😭 (classic).

⚠️ lowkey important: this must be called while holding the devlink instance lock. fr fr, do not skip the lock, that's not a vibe 💀

Snapshot IDs are tracked using an xarray which stores the number of users of the snapshot id 🌈. Like, the xarray is keeping count of how many people are holding onto each id — very main character bookkeeping energy 👑

Also note 📝: the caller of this function counts as a 'user', in order to avoid race conditions. So yes, you're part of the count the second you call her 🙈. That means the caller must release its hold on the snapshot by using devlink_region_snapshot_id_put — like, you gotta give the id back when you're done, don't just ghost it 💀💖

✨ SEE ALSO ✨

Kernel file ./net/devlink/region.c plus all these besties 🦋:

and that's the page, bestie 💜 this was brought to you by ./net/devlink, October 2026 — periodt, go forth and manage your snapshot ids responsibly 🧚‍♀️✨🎀

📦 where this page comes from