ib_cq_pool_get(9)

📦 version 7.3.1, in nixpkgs unstable

💅 ib_cq_pool_get — like, the Kernel API Manual page, bestie

okay so bestie, welcome to ib_cq_pool_get(9) from the Kernel API Manual ✨ this is the one where the kernel goes shopping for a completion queue for you and honestly? she ate. no cap.

🌸 NAME 🌸

ib_cq_pool_get — finds the least used completion queue that matches a given cpu hint (or, for wild card affinity, just the least used one) and also fits nr_cqe. literally the matchmaker of CQs 🫶

✨ SYNOPSIS ✨

struct ib_cq * ib_cq_pool_get (struct ib_device *dev , unsigned int
nr_cqe , int comp_vector_hint , enum ib_poll_context poll_ctx );

like, that's the whole vibe. four args, one cq pointer back. periodt. 💅

🎀 ARGUMENTS 🎀

  • dev — the rdma device, aka the main character of this whole situation 👑
  • nr_cqe — number of needed cqe entries. like how many seats you're reserving 🎟️
  • comp_vector_hint — completion vector hint (-1) for the driver to assign a comp vector based on internal counter. yes, -1 is the "idk you pick for me" option 🙈
  • poll_ctx — cq polling context. the vibe of the polling, if you will 🌈

🦋 DESCRIPTION 🦋

so here's the tea ☕ — she finds a cq that satisfies the comp_vector_hint and nr_cqe requirements, and then claims entries in it for us. iconic behavior, honestly.

and if there's no available cq? no worries, she just allocates a brand new cq with those requirements and adds it to the device pool. she's resourceful like that 💖

⚠️ warning bestie: IB_POLL_DIRECT cannot be used for shared cqs, so it is not a valid value for poll_ctx. do not try it. i'm begging you. 😭

🌟 SEE ALSO 🌟

go read those too, they're all part of the same friend group fr fr 🧚

and that's the page!! 🎉 this little gem lives in ./drivers/infiniband/core, dated October 2026, and its whole name is ib_cq_pool_get(9). mwah 💋 go get yourself a completion queue, you deserve it. ⭐🍓

📦 where this page comes from