Hi Bart, On Wed, 1 Feb 2017 16:58:17 -0800, Bart Van Assche wrote: > The most important changes in this patch series are: > * ABORT and LUN RESET handling is made synchronous. I occasionally hit a lockup while running the libiscsi MultipathIO.Reset test against a scsi-target-for-v4.11 iSCSI target, which I haven't seen on mainline (yet): rapido1:/# ps -eo args,wchan ... 619 [iscsi_ttx] iscsit_cause_connection_reinstatement 620 [iscsi_trx] transport_wait_for_tasks 624 [iscsi_trx] transport_wait_for_tasks 627 [kworker/u4:2] core_tmr_lun_reset 632 [iscsi_ttx] iscsi_target_tx_thread 633 [iscsi_trx] wait_woken 636 [kworker/u4:3] worker_thread 637 [kworker/u4:4] worker_thread 640 ps -eo pid,args,wchan - rapido1:/# cat /proc/627/stack [<ffffffffa0012783>] core_tmr_lun_reset+0x1f3/0x740 [target_core_mod] [<ffffffffa0015341>] target_tmr_work+0xe1/0x130 [target_core_mod] [<ffffffff810563e9>] process_one_work+0x129/0x300 [<ffffffff810566e6>] worker_thread+0x126/0x480 [<ffffffff8105b877>] kthread+0xf7/0x130 [<ffffffff815452d9>] ret_from_fork+0x29/0x40 [<ffffffffffffffff>] 0xffffffffffffffff rapido1:/# cat /proc/619/stack [<ffffffffa00663d9>] iscsit_cause_connection_reinstatement+0x99/0xe0 [iscsi_target_mod] [<ffffffffa00781d9>] iscsit_stop_session+0x119/0x1c0 [iscsi_target_mod] [<ffffffffa0078ec3>] iscsit_logout_post_handler+0xa3/0x200 [iscsi_target_mod] [<ffffffffa00790c4>] iscsit_response_queue+0xa4/0x690 [iscsi_target_mod] [<ffffffffa0077ec0>] iscsi_target_tx_thread+0x100/0x1b0 [iscsi_target_mod] [<ffffffff8105b877>] kthread+0xf7/0x130 [<ffffffff815452d9>] ret_from_fork+0x29/0x40 [<ffffffffffffffff>] 0xffffffffffffffff The initiator blocks waiting for a response to the TMF LUN RESET req, but never gets one. Cheers, David -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
