propagate_umount(9)

💅 NAME 💅

propagate_umount — bestie, this is the kernel function (from the Kernel API Manual, hey girl hey) that applies propagation rules to the set of mounts for umount(). It's giving "cleanup crew with a clipboard" energy ✨

---

🦋 SYNOPSIS 🦋

void propagate_umount (struct list_head *set );

That's it. That's the whole signature. No cap. 🫶

---

🎀 ARGUMENTS 🎀

• set — the list of mounts to be unmounted. That's your starting squad, your little guest list of mounts that are already getting the boot 👢

---

✨ DESCRIPTION ✨

Okay so here's the vibe: propagate_umount() collects all the mounts that receive propagation from the mount in set and have no obstacles to being unmounted. Iconic behaviour, honestly.

Then — and this is the main character moment — it adds these additional mounts to the set. So the guest list grows. Lowkey a group chat adding people to the plan 😭

🚨 real talk: see Documentation/filesystems/propagate_umount.txt if you do anything in this area. Like, anything. Do not freestyle this one, bestie, I'm begging 🙈

Locks held: mount_lock (write_seqlock), namespace_sem (exclusive). Yes, both. Exclusive namespace_sem is the big one — respect the lock order or the kernel will simply end you 💀

---

🔥 SEE ALSO 🔥

• Kernel file ./fs/pnode.c propagate_mnt(9) — the sibling function, the origin story 🧚
• propagation_would_overmount(9) — the vibe check before you overmount, iykyk ⭐

---

Slay, that's the whole page 💖 brought to you by source ./fs, dated October 2026, from the propagate_umount(9) fam — go forth and unmount responsibly 👑🌈