inet_csk_clone_lock(9)

inet_csk_clone_lock(9) — clone an inet socket, and lock its clone 💖

okay bestie, this is the Kernel API Manual entry for inet_csk_clone_lock, and it’s literally the “make a clone and lock it” moment. We’re keeping every detail, no cap.

NAME ✨

inet_csk_clone_lock — clone an inet socket, and lock its clone 🎀

SYNOPSIS 🦋

struct sock * inet_csk_clone_lock (const struct sock *sk , const struct
       request_sock *req , const gfp_t priority );

ARGUMENTS 🎀

• sk — the socket to clone 💅
• req — request_sock
• priority — for allocation (GFP_KERNEL, GFP_ATOMIC, etc)

DESCRIPTION 🔥

The caller must unlock the socket even in the error path — yes, bestie, even when things go sideways 😭. The original spells it out as bh_unlock_sock(newsk), and that cleanup is not optional. It’s giving mandatory-unlock energy, fr fr.

no skipping the unlock, even on the error path — that’s the whole caveat 👑

SEE ALSO 🌈

• Kernel file ./net/ipv4/inet_connection_sock.c
• inet_sk_get_local_port_range(9)

okay bye bestie 🫶 this one comes from source ./net/ipv4, dated October 2026, and the page is inet_csk_clone_lock(9) ✨💅