icc_set_bw(9)

📦 version 7.3.1, in nixpkgs unstable

💅 icc_set_bw(9) — the kernel API manual, but make it make sense

okay bestie, welcome to the icc_set_bw page 💖 this one lives in the Kernel API Manual and it's literally the glow-up your interconnect path has been waiting for. no cap.

✨ NAME ✨

icc_set_bw — set bandwidth constraints on an interconnect path 🎀

🌸 SYNOPSIS 🌸

int icc_set_bw (struct icc_path *path , u32 avg_bw , u32 peak_bw );

🦋 ARGUMENTS 🦋

  • path — the interconnect path. that's your lil route between two endpoints, the main character of this whole function ✨
  • avg_bw — average bandwidth, measured in kilobytes per second 📏
  • peak_bw — peak bandwidth, also in kilobytes per second 🔥 (yes, the spicy one)

👑 DESCRIPTION 👑

so here's the vibe: this function is what an interconnect consumer calls to say "hey, these are my needs" — in terms of bandwidth — for a path it already requested earlier between two endpoints. you're not creating anything here, you're just expressing your truth 💅

the requests get aggregated, and then each node along the way is updated accordingly. the entire path is locked by a mutex to make sure the whole set actually completes — nobody's cutting in mid-update, periodt. 🔒

quick heads-up bestie: path can totally be NULL — that happens when the "interconnects" DT properties are missing. in that case, it just means no constraints will be set. no drama, no constraints, we move on. 🫶

returns 0 on success, or an appropriate error code otherwise. that's it, that's the return 😭

🌈 SEE ALSO 🌈

the kernel file lives at ./drivers/interconnect/core.c — go say hi 👋 and while you're there, here's the whole friend group:

okay that's the whole page, you ate that 💅✨ this one's from ./drivers/interconnect, dated October 2026 — see you in the next manpage, bestie, mwah 🎀💖

📦 where this page comes from