aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorChristof Schmitt <christof.schmitt@de.ibm.com>2009-07-13 15:06:10 +0200
committerJames Bottomley <James.Bottomley@HansenPartnership.com>2009-07-30 08:49:57 -0500
commitcbf1ed0264da104573458aedc220ebfcd02567f6 (patch)
treee3a8e5f815fd22c556a61189c0d3e0d0658fa87d
parent85600f7f8370fe5b4be0debd8b401de7986b52ae (diff)
[SCSI] zfcp: Recover from stalled outbound queue
Depending on interruptions on some storage systems, the complete channel can stall which looks like an outbound queue stall to Linux. When trying to acquire a free SBAL for a non-SCSI command, zfcp waits for 5 seconds for a free slot to appear. This is the right place to detect a queue stall: If the wait times out, we assume a stalled queue and try to recover this. The overall strategy should be to trigger the erp from specific events, and not try an overall escalation from one failed port to a full-blown queue recovery. If we manage to send a command, the status codes for this command or a timeout will trigger the right follow-on actions. Reviewed-by: Swen Schillig <swen@vnet.ibm.com> Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
-rw-r--r--drivers/s390/scsi/zfcp_fsf.c5
1 files changed, 4 insertions, 1 deletions
diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c
index 0c24695a53c..b7e48844056 100644
--- a/drivers/s390/scsi/zfcp_fsf.c
+++ b/drivers/s390/scsi/zfcp_fsf.c
@@ -670,8 +670,11 @@ static int zfcp_fsf_req_sbal_get(struct zfcp_adapter *adapter)
zfcp_fsf_sbal_check(adapter), 5 * HZ);
if (ret > 0)
return 0;
- if (!ret)
+ if (!ret) {
atomic_inc(&adapter->qdio_outb_full);
+ /* assume hanging outbound queue, try queue recovery */
+ zfcp_erp_adapter_reopen(adapter, 0, "fsrsg_1", NULL);
+ }
spin_lock_bh(&adapter->req_q_lock);
return -EIO;