ibmvscsis_post_disconnect(9)

💅 ibmvscsis_post_disconnect(9) — the Kernel API Manual page you didn't know you needed 🎀

okay so bestie, we're in the Kernel API Manual today and honestly? this one's a little function with main character energy. it schedules a disconnect. that's it. that's the vibe. let's get into it ✨

---

✨ NAME ✨

ibmvscsis_post_disconnect — Schedule the disconnect 🦋

like, it posts the disconnect. it doesn't do the whole thing itself, it just puts it on the schedule. very "i've added it to the calendar, don't @ me" of it.

---

🌸 SYNOPSIS 🌸

void ibmvscsis_post_disconnect (struct scsi_info *vscsi , uint new_state
, uint flag_bits );

returns void, meaning it gives you nothing back, and honestly? respect. no return value, no pressure, no expectations. it just does its thing. 💅

---

🫶 ARGUMENTS 🫶

• vscsi — Pointer to our adapter structure. literally our adapter. it's giving possessive, it's giving main character, and that's exactly what it is 🎀
• new_state — State to move to after disconnecting. so like, where are we going after the disconnect? this tells it. destination: unknown but specified 💖
• flag_bits — Flags to turn on in adapter structure. you're flipping switches on the adapter, slay ⭐

---

🔥 DESCRIPTION 🔥

so here's the tea, bestie: if it's already been scheduled, then see if we need to "upgrade" the new state (if the one passed in is more "severe" than the previous one).

read that again. upgrade. quotes around it and everything. the manpage is literally putting "upgrade" in scare quotes because severity going up is not exactly a glow-up, y'know? 😭

PRECONDITION: interrupt lock is held 🔒

no cap, this is a big one. the interrupt lock is held when you call this. not "should be", not "maybe". is held. don't be the one who forgot. i'm begging 🙈

---

👑 SEE ALSO 👑

Kernel file ./drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c — go read it, iykyk 💜

and the whole friend group, in the order the page gives them:

connection_broken(9), ibmvscsis_unregister_command_q(9), ibmvscsis_delete_client_info(9), ibmvscsis_free_command_q(9), ibmvscsis_cmd_q_dequeue(9), ibmvscsis_send_init_message(9), ibmvscsis_check_init_msg(9), ibmvscsis_disconnect(9), ibmvscsis_handle_init_compl_msg(9), ibmvscsis_handle_init_msg(9), ibmvscsis_init_msg(9), ibmvscsis_establish_new_q(9), ibmvscsis_reset_queue(9), ibmvscsis_free_cmd_resources(9), ibmvscsis_ready_for_suspend(9), ibmvscsis_trans_event(9), ibmvscsis_poll_cmd_q(9), ibmvscsis_free_cmd_qs(9), ibmvscsis_get_free_cmd(9), ibmvscsis_adapter_idle(9), ibmvscsis_copy_crq_packet(9), ibmvscsis_adapter_info(9), ibmvscsis_cap_mad(9), ibmvscsis_process_mad(9), srp_snd_msg_failed(9), ibmvscsis_send_messages(9), ibmvscsis_mad(9), ibmvscsis_login_rsp(9), ibmvscsis_srp_login_rej(9), ibmvscsis_srp_login(9), ibmvscsis_srp_i_logout(9), ibmvscsis_ping_response(9), ibmvscsis_parse_command(9), ibmvscsis_parse_cmd(9), ibmvscsis_parse_task(9), ibmvscsis_service_wait_q(9), ibmvscsis_enable_change_state(9), ibmvscsis_create_command_q(9), ibmvscsis_destroy_command_q(9), srp_build_response(9), ibmvscsis_handle_crq(9)

like, the entire extended friend group showed up to this SEE ALSO. forty-two of them. iconic behavior, no notes 🌈

---

aaaand that's the page! 💖 big lil sign-off moment: this one lives in ./drivers/scsi/ibmvscsi_tgt and it's dated October 2026 — so mark your calendars bestie, we're reading kernel docs from the future 🦋✨ periodt.