--- /dev/null
+From: Gerald Schaefer <geraldsc@de.ibm.com>
+Subject: zfcp: Recover from stalled outbound queue
+References: bnc#518291,LTC#54460
+
+Symptom: In some error scenarious, the channel can stall. To
+ Linux this looks like a stall in the outbound queue.
+ Linux has to recover this.
+Problem: There is no trigger to recover the queues if the outbound
+ queue stalls.
+Solution: When trying to issue a non-SCSI command, zfcp waits for
+ 5 seconds to get a free SBAL in the queue. Use this timeout
+ as a trigger: If after 5 seconds there is no free SBAL,
+ assume a stalled outbound queue and trigger recovery.
+
+Acked-by: John Jolly <jjolly@suse.de>
+---
+ drivers/s390/scsi/zfcp_dbf.c | 2 +-
+ drivers/s390/scsi/zfcp_fsf.c | 5 ++++-
+ 2 files changed, 5 insertions(+), 2 deletions(-)
+
+--- a/drivers/s390/scsi/zfcp_fsf.c 2009-06-23 14:24:48.000000000 +0200
++++ b/drivers/s390/scsi/zfcp_fsf.c 2009-06-23 14:26:00.000000000 +0200
+@@ -668,8 +668,11 @@ static int zfcp_fsf_req_sbal_get(struct
+ 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, 77, NULL);
++ }
+
+ spin_lock_bh(&adapter->req_q_lock);
+ return -EIO;
+--- a/drivers/s390/scsi/zfcp_dbf.c 2009-06-23 14:20:41.000000000 +0200
++++ b/drivers/s390/scsi/zfcp_dbf.c 2009-06-23 14:25:50.000000000 +0200
+@@ -567,7 +567,7 @@ static const char *zfcp_rec_dbf_ids[] =
+ [74] = "unit recovery follow-up after successful port recovery",
+ [75] = "port recovery retry after failure",
+ [76] = "unit recovery retry after failure",
+- [77] = "",
++ [77] = "adapter reopen from queue stall",
+ [78] = "duplicate request id",
+ [79] = "link down",
+ [80] = "exclusive read-only unit access unsupported",