irq_migrate_...off_this_cpu(9)
π irq_migrate_all_off_this_cpu
okay bestie, welcome to the Kernel API Manual β yes, that's the whole vibe, we're in the kernel docs now and it's giving β¨ deep magic β¨. Sit down, put your hair up, and let's talk about getting interrupts off a CPU that's clocking out. π
πΈ NAME πΈ
irq_migrate_all_off_this_cpu β migrating IRQs away from an offline CPU, no cap π
π¦ SYNOPSIS π¦
void irq_migrate_all_off_this_cpu (void );
π ARGUMENTS π
voidβ no arguments, babe. zero. nothing. it takes nothing and it knows it. π
β¨ DESCRIPTION β¨
So here's the tea β: the current CPU has already been marked offline. Like, it's out, it's done, it's not coming to the function.
This function's whole job is to migrate IRQs off this CPU. Get them gone. Move them somewhere that still has main character energy. π₯
And if the affinity settings do not allow other CPUs? Then the function forces them onto any available CPU. Yeah. It does not ask. It does not negotiate. It just puts them somewhere π«Ά
no cap, that's the entire description β but it's an important one, so we're being loud about it π
π NOTE π
we must iterate over all IRQs, whether they have an attached action structure or not, as we need to get chained interrupts too.
like β all of them. even the ones with no action structure attached. because chained interrupts exist and we need them, iykyk π
π SEE ALSO π
- Kernel file
./kernel/irq/cpuhotplug.cπ irq_affinity_online_cpu(9)β the other one, go read it too, bestie π
lowkey aesthetic moment: this whole page lives under the ./kernel/irq section, and the docs are dated October 2026, so yeah β we're reading the future, periodt β
stay hydrated, go read cpuhotplug.c, and remember: if the affinity settings don't let you move, someone will just force you onto another CPU anyway ππ bye bestie π§
π¦ where this page comes from
- nixpkgs unstable: linux-manual 7.3.1
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms