iscsi_check_...restrictions(9)
π
iscsi_check_...restrictions (9) β like, the Kernel API Manual page, bestie
okay so real talk, this one lives in the Kernel API Manual and it's genuinely not as scary as the name looks. the name got truncated with ... because man page titles have a character limit and honestly? relatable. anyway β we're checking whether a task got caught up in a TMF. let's go π
β¨ NAME β¨
iscsi_check_tmf_restrictions β literally just checking if a task is affected by TMF, no cap π
πΈ SYNOPSIS πΈ
int iscsi_check_tmf_restrictions (struct iscsi_task *task , int opcode
);
yes the line break is exactly like that, don't @ me, copy it verbatim π«Ά
π¦ ARGUMENTS π¦
taskβ the iscsi task in question. the main character of this function, basically πopcodeβ the opcode you're checking for. like, which one are we vibe-checking here π₯
β DESCRIPTION β
so during a TMF, a task has to get checked to see if it's affected. that's the whole gig. β¨
the good news? all the unrelated I/O can just pass straight through β no drama, no notes, slay. but I/O heading to the affected LUN? that should be restricted. periodt. π
and here's the tea π:
- if
'fast_abort'is set, we won't be sending any I/O to the affected LUN. like, none. zero. gone. π - otherwise, the target is waiting for all the TTTs to be completed, so we have to send all outstanding Data-Out PDUs to the target. no skipping, bestie β every single one gets sent. π«Ά
ngl this is one of those "the details matter" situations β miss the 'fast_abort' branch and you're having a bad time π
π SEE ALSO π
Kernel file ./drivers/scsi/libiscsi.c plus literally the whole friend group:
iscsi_prep_data_out_pdu(9)iscsi_prep_scsi_cmd_pdu(9)iscsi_free_task(9)__iscsi_put_task(9)iscsi_complete_task(9)iscsi_complete_scsi_task(9)iscsi_alloc_mgmt_task(9)iscsi_send_mgmt_task(9)iscsi_scsi_cmd_rsp(9)iscsi_data_in_rsp(9)iscsi_nop_out_rsp(9)iscsi_itt_to_task(9)__iscsi_complete_pdu(9)iscsi_itt_to_ctask(9)iscsi_requeue_task(9)iscsi_data_xmit(9)iscsi_suspend_queue(9)iscsi_suspend_tx(9)iscsi_suspend_rx(9)iscsi_conn_unbind(9)iscsi_eh_session_reset(9)iscsi_eh_target_reset(9)iscsi_eh_recover_target(9)iscsi_host_add(9)iscsi_host_alloc(9)iscsi_host_remove(9)iscsi_session_setup(9)iscsi_session_free(9)iscsi_session_teardown(9)iscsi_conn_setup(9)iscsi_conn_teardown(9)
that's the whole group chat, all 31 of them, iykyk π
and that's the page!! π§ sign-off time: this one's from the source dir ./drivers/scsi, dated October 2026 β so go forth, check your TMF restrictions, and remember: unrelated I/O stays unbothered, affected LUNs get restricted. mwah ππ
π¦ where this page comes from
- nixpkgs unstable: linux-manual 7.3.1
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms