rdmacg_uncharge(9)

📦 version 7.3.1, in nixpkgs unstable

💅 rdmacg_uncharge — the kernel API manual, bestie edition

Okay so real talk, this whole page is about rdmacg_uncharge, which is a function from the Kernel API Manual. Yes, that Kernel API Manual. Pull up a chair 🎀

✨ NAME ✨

rdmacg_uncharge — hier<a>chically uncharge rdma resource count

Not me being the one who gives back the RDMA resources 💀 it's literally the inverse of charging, and she does it hierarchically, up the whole family tree.

🌸 SYNOPSIS 🌸

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

Notice it returns void? She gives nothing back on the return-value front. She's just out here uncharging. Periodt. ✨

🦋 ARGUMENTS 🦋

  • cg — pointer to cg to uncharge and all parents in hierarchy 👑 this is a whole family situation, not a solo mission. Every ancestor gets visited too.
  • device — pointer to the rdmacg device. That's the device bestie, what else.
  • index — index of the resource to uncharge in the cgroup in the given resource pool. Like, which resource, specifically, from which pool. Precision queen 💅

👑 SEE ALSO 👑

The kernel file ./kernel/cgroup/rdma.c and its friends — literally the whole squad, no cap:

Fr fr, if you're poking around in here you probably also want the charging side of the story (rdmacg_try_charge(9)), because what is uncharging without charging 😭 they're a matched set, iykyk.

And that's the vibe check 🧚 This icon of a page lives under ./kernel/cgroup, dated October 2026, and yes she's still rdmacg_uncharge(9) — mwah, slay, go read the source bestie 🌈💖

📦 where this page comes from