Hello Guys,
fnic uses the following way to walk scsi commands in failure handling,
which is obvious wrong, because caller of scsi_host_find_tag has to
guarantee that the tag is active.
for (tag = 0; tag < fnic->fnic_max_tag_id; tag++) {
...
sc = scsi_host_find_tag(fnic->lport->host, tag);
...
}
Fix the issue by using scsi_host_busy_iter() to walk request/scsi_command.
Hannes has posted similar patch[1] before, however which depends on the scsi
reserved command patchset, so post them out in one standalone patchset,
another difference is that this one is splited to 5 single patches for
easy review.
[1] https://marc.info/?l=linux-scsi&m=161400059528859&w=2
V2:
- use scsi_host_busy_iter
Ming Lei (5):
scsi: fnic: use scsi_host_busy_iter in fnic_terminate_rport_io
scsi: fnic: use scsi_host_busy_iter in fnic_clean_pending_aborts
scsi: fnic: use scsi_host_busy_iter in fnic_cleanup_io
scsi: fnic: use scsi_host_busy_iter in fnic_rport_exch_reset
scsi: fnic: use scsi_host_busy_iter in fnic_is_abts_pending
drivers/scsi/fnic/fnic_scsi.c | 924 ++++++++++++++++++----------------
1 file changed, 484 insertions(+), 440 deletions(-)
--
2.29.2
