skb_clone(9)

📦 version 7.3.1, in nixpkgs unstable

🦋 NAME

skb_clone — duplicate an sk_buff 💅

hi bestie, welcome to the skb_clone(9) page, straight out of the Kernel API Manual ✨ yes, the whole thing is about making a twin of a socket buffer, and it's lowkey iconic how much drama one struct can hold 🌸

🎀 SYNOPSIS

struct sk_buff * skb_clone (struct sk_buff *skb , gfp_t gfp_mask );

that's the whole signature, no cap 👑

💖 ARGUMENTS

  • skb — the buffer you're cloning, aka the one getting a twin 🫶
  • gfp_mask — the allocation priority, like how hard you're willing to try to get memory 😭

🔥 DESCRIPTION

okay so, you want a duplicate of an sk_buff? say less 💁‍♀️ The new one is not owned by a socket — it's out here living its own life, no socket holding its hand 🙈

The way it works is genuinely so cute: both copies share the same packet data but not structure. So they're matching outfits on the data but they each have their own personality in the structure department 💅 slay.

The new buffer comes out with a reference count of 1 — brand new main character energy, zero baggage 👑

And about failure? If the allocation fails, the function returns NULL. Otherwise you get the new buffer back. That's it. That's the whole vibe 🌈

⚠️ bestie, listen to me: if this function is called from an interrupt, gfp_mask must be GFP_ATOMIC. not "should", not "probably", must 💀 don't be the one who finds out why.

👑 SEE ALSO

It's giving family reunion, fr fr — the kernel file is ./net/core/skbuff.c and here's everyone at the party 🎉:

and that's the page, bestie 🍓 mwah 💋

little sign-off moment: this one comes from ./net/core, dated October 2026, and the page is skb_clone(9) — go forth and clone responsibly ✨🧚‍♀️

📦 where this page comes from