ib_send_cm_dreq(9)

📦 version 7.3.1, in nixpkgs unstable

💅 ib_send_cm_dreq — from the Kernel API Manual (yes bestie, the kernel one)

okay so picture this: you've got a connection, it's been cute, it's been doing its thing, and now it's time to say bye. This little icon right here, ib_send_cm_dreq, is literally the "we need to talk" text — it sends a disconnection request for a connection that already exists. No cap, that's the whole job. 💔✨

✨ SYNOPSIS ✨

int ib_send_cm_dreq (struct ib_cm_id *cm_id , const void *private_data ,
u8 private_data_len );

the way this signature just sits there being all three-argument and tidy 🎀 returns an int, takes your connection id, your optional lil private data blob, and how big that blob is. slay.

🫶 ARGUMENTS 🫶

  • cm_id — the connection identifier tied to the connection you're releasing. it's giving "which one of my connections am I breaking up with" energy. main character of this call, honestly. 👑
  • private_data — optional user-defined private data that gets sent along with the disconnection request message. optional!! you don't have to send anything. but if you do, it rides shotgun with the dreq. 🦋
  • private_data_len — the size of that private data buffer, in bytes. it's a u8, so keep it smol and keep it honest. 🌸

🌈 SEE ALSO 🌈

Kernel file ./include/rdma/ib_cm.h — go look at it, iykyk 📖 — plus this whole friend group right here:

like a whole group chat of connection-manager functions. you send the dreq, then you go peek at ib_send_cm_drep(9) to see the other side of the breakup. 💀💖

lowkey reminder bestie: private_data is optional and private_data_len is its byte count — don't be out here passing a length that doesn't match your buffer, that's not the vibe. 🙈

and that's the page!! 🎀 sign-off time: this one lives over in ./include/rdma, dated October 2026, and it's filed under ib_send_cm_dreq(9) in the Kernel API Manual. obsessed with her, honestly. periodt. ✨💅🫶

📦 where this page comes from