fcoe_percpu_clean(9)

📦 version 7.3.1, in nixpkgs unstable

💅 fcoe_percpu_clean — from the Kernel API Manual, section 9, bestie

okay so real talk, this page is about fcoe_percpu_clean, and yes it's a kernel API manual thing (section 9, the fancy driver-internals section, iykyk) ✨

🌸 NAME 🌸

fcoe_percpu_clean — literally clears out all the pending skbs for one local port 🧹💖

✨ SYNOPSIS ✨

void fcoe_percpu_clean (struct fc_lport *lport );

like, that's it. that's the whole signature. no return value, no drama, just vibes 💅

🎀 ARGUMENTS 🎀

  • lport — the local port whose skbs are getting cleared out. she's the main character here, everyone else is just supporting cast 🫶

🦋 DESCRIPTION 🦋

okay so first thing first, and this is not optional, bestie: this function must be called with fcoe_create_mutex held to single-thread completion. like, no cap. hold the mutex. fr fr. 🔒

what it does: it flushes the pending skbs by flushing the work item for each CPU. the work item on each possible CPU gets flushed — and here's the why, because it's lowkey important — because we may have used the per-CPU struct of an offline CPU. so yeah. possible CPUs, all of them, not just the online ones. she's thorough like that 💀✨

🌈 SEE ALSO 🌈

kernel file ./drivers/scsi/fcoe/fcoe.c and honestly the whole friend group:

like I said. the whole friend group. every single one made the list 👑💜

🫶 reminder: mutex held = good. mutex not held = chaos. don't do chaos.

and that's the tea, bestie 🍓 this one's from ./drivers/scsi/fcoe, dated October 2026 — mwah, iconic, no notes 💋✨🎀

📦 where this page comes from