kernel_accept(9)

💖 NAME 💖

kernel_accept — accept a connection (kernel space) ✨

okay so bestie, this is kernel_accept(9) from the Kernel API Manual, and it's literally the kernel-space way to accept a connection. no cap. 🫶

---

🌸 SYNOPSIS 🌸

int kernel_accept (struct socket *sock , struct socket **newsock , int
       flags );

---

🎀 ARGUMENTS 🎀

• sock — the listening socket, aka the one out here waiting for a connection to slide into its DMs 💌
• newsock — the new connected socket, aka the fresh little connection that shows up after accept slays ✨
• flags — the flags. yes, that's the whole description. iconic brevity, honestly 💅

---

🦋 DESCRIPTION 🦋

okay so here's the vibe check: flags must be SOCK_CLOEXEC, SOCK_NONBLOCK, or 0. those are the only choices, bestie — no improvising, no "oops I made up a flag" energy 🙈

if it fails, newsock is guaranteed to be NULL. like, that's the promise. no dangling fake socket, periodt. 🔥

it returns 0 or an error. that's it. that's the whole return situation 💀

---

🌈 SEE ALSO 🌈

Kernel file ./net/socket.c — and the whole extended friend group, iykyk:

• move_addr_to_kernel(9)
• move_addr_to_user(9)
• sock_read_xattr(9)
• sock_alloc_file(9)
• sock_from_file(9)
• sockfd_lookup(9)
• sock_alloc(9)
• sock_release(9)
• sock_sendmsg(9)
• kernel_sendmsg(9)
• sock_recvmsg(9)
• kernel_recvmsg(9)
• sock_create_lite(9)
• __sock_create(9)
• sock_create(9)
• sock_create_kern(9)
• sock_register(9)
• sock_unregister(9)
• kernel_bind(9)
• kernel_listen(9)
• kernel_connect(9)
• kernel_getsockname(9)
• kernel_getpeername(9)
• kernel_sock_shutdown(9)
• kernel_sock_ip_overhead(9)

---

okay bye bestie 🫶 — source/version is ./net and the date is October 2026, so that's the cute lil sign-off ✨💖