STRUCT XFRM_IPTFS_DATA(9)

🦋 struct xfrm_iptfs_data(9) 🦋

okay so this is a page from the Kernel API Manual (section 9, bestie, kernel API land) and honestly? it's giving "the blueprint for one IPTFS tunnel's whole personality." buckle up 💖

---

✨ NAME ✨

struct xfrm_iptfs_data — mode specific xfrm state. literally what it says on the tin: all the state that only makes sense for this one xfrm mode. no cap 🎀

---

🌸 SYNOPSIS 🌸

struct xfrm_iptfs_data {
    struct xfrm_iptfs_config cfg;
    struct xfrm_state *x;
    struct sk_buff_head queue;
    u32 queue_size;
    u32 ecn_queue_size;
    u64 init_delay_ns;
    struct hrtimer iptfs_timer;
    time64_t iptfs_settime;
    u32 payload_mtu;
    bool w_seq_set;
    u64 w_wantseq;
    struct skb_wseq *w_saved;
    u32 w_savedlen;
    spinlock_t drop_lock;
    struct hrtimer drop_timer;
    u64 drop_time_ns;
    struct sk_buff *ra_newskb;
    u64 ra_wantseq;
    u8 ra_runt[6];
    u8 ra_runtlen;
};

like, that's the whole struct, every field, nothing skipped, periodt 💅

---

👑 Members 👑

• cfg — the IPTFS tunnel config. the settings bestie 🎀
• x — the owning SA (xfrm_state). this struct's main character.
• queue — queued user packets that are waiting to go out. the to-send pile ✨
• queue_size — how many octets are sitting on queue (that's the sum of the packet sizes). not a packet count, an octet count, ok? 💖
• ecn_queue_size — octets up there that carry the ECN mark.
• init_delay_ns — nanoseconds to wait before sending the initial IPTFS packet. patience, gorgeous, patience 🌸
• iptfs_timer — the output timer.
• iptfs_settime — the time the output timer was set.
• payload_mtu — max payload size.
• w_seq_set — true after the first seq has been received.
• w_wantseq — we're waiting for this seq number as the next one to process (in order, always in order, we love an in-order queen 👑).
• w_saved — the saved buf array (that's the reorder window).
• w_savedlen — the saved len (not the size, bestie — len. w_savedlen. not the size 😭)
• drop_lock — lock protecting the reorder queue. door needs a lock, obviously 🫶
• drop_timer — timer for considering the next packet lost.
• drop_time_ns — timer intervan in nanoseconds (that's the page's spelling, not mine, iykyk 💀).
• ra_newskb — the new pkt being reassembled.
• ra_wantseq — expected next sequence for reassembly.
• ra_runt — last pkt bytes from the very end of the last skb. it's u8 ra_runt[6] fr fr.
• ra_runtlen — size of ra_runt.

---

🔥 SEE ALSO 🔥

kernel file ./net/xfrm/xfrm_iptfs.c plus this whole friend group:

IPTFS_DEFAULT_DROP_TIME_USECS(9), IPTFS_DEFAULT_REORDER_WINDOW(9), IPTFS_DEFAULT_INIT_DELAY_USECS(9), IPTFS_DEFAULT_MAX_QUEUE_SIZE(9), struct xfrm_iptfs_config(9), iptfs_alloc_skb(9), iptfs_skb_head_to_frag(9), struct iptfs_skb_frag_walk(9), iptfs_skb_prepare_frag_walk(9), iptfs_skb_can_add_frags(9), iptfs_skb_add_frags(9), iptfs_pskb_add_frags(9), iptfs_pskb_extract_seq(9), iptfs_input_save_runt(9), __iptfs_iphlen(9), __iptfs_iplen(9), iptfs_complete_inner_skb(9), iptfs_reassem_abort(9), iptfs_reassem_done(9), iptfs_reassem_cont(9), iptfs_input_ordered(9), iptfs_drop_timer(9), iptfs_input(9), iptfs_enqueue(9), iptfs_copy_create_frag(9), iptfs_copy_create_frags(9), iptfs_first_skb(9), iptfs_encap_add_ipv4(9), iptfs_encap_add_ipv6(9), iptfs_prepare_output(9), __iptfs_get_inner_mtu(9), iptfs_get_inner_mtu(9), iptfs_user_init(9)

---

ngl every single one of those is a real thing you can look up, i'm not making them up, that would be so not slay of me 🙈 💅

okay bye bestie, go read the source, it's iconic 🌈⭐

— from ./net/xfrm, dated October 2026. catch you in the kernel tree, mwah 🍓🧚💜