rdmacg_event_locked(9)

📦 version 7.3.1, in nixpkgs unstable

💅 rdmacg_event_locked — like, the Kernel API Manual one, bestie

okay so we're in the Kernel API Manual today and the star of the show is rdmacg_event_locked — say that five times fast, i actually can't 😭

✨ NAME ✨

rdmacg_event_locked — fire an event when a resource allocation goes over the limit 🔥

🌸 SYNOPSIS 🌸

void rdmacg_event_locked (struct rdma_cgroup *cg , struct rdma_cgroup
*over_cg , struct rdmacg_device *device , enum rdmacg_resource_type
index );

like yes bestie it returns void, it's giving "i just do my thing and leave" 💅

🎀 ARGUMENTS 🎀

  • cg — the cgroup that's doing the requesting, aka the one asking for stuff 🫶
  • over_cg — the cgroup whose limit got exceeded, rip to her honestly 💀
  • device — the rdma device, the whole reason we're here ✨
  • index — the resource type index, like which resource we're talking about 🦋

🌈 DESCRIPTION 🌈

okay so the one thing you cannot skip: this must be called under rdmacg_mutex. no cap. that's the vibe check and you have to pass it 🔒

once you're in there, here's the whole little routine she runs:

  • updates the event counters in the resource pools of cg and over_cg 👑
  • propagates hierarchical max events from over_cg (including itself, yes, herself too, main character energy) upward ⬆️
  • notifies userspace via cgroup_file_notify 📣

that's it, that's the whole function, and honestly? she ate 💖

🧚 SEE ALSO 🧚

Kernel file ./kernel/cgroup/rdma.c

like go read those too, iykyk 🍓

fr fr: if you're not holding rdmacg_mutex when you call this, you're doing it wrong and i will be judging you 🙈

and that's the whole page bestie!! sourced from ./kernel/cgroup, dated October 2026, all wrapped up in rdmacg_event_locked(9) — mwah, slay, periodt 💜🎀✨

📦 where this page comes from