[SCSI] zfcp: Trigger logging in the FCP channel on qdio error conditions

Exploit the cio siosl function to trigger logging in the FCP channel
on qdio error conditions. Add a helper function in zfcp_qdio to ensure
that tracing is only triggered once before calling qdio_shutdown.

Trigger in zfcp for hardware logs are:
 - timeout for FSF requests to the FCP channel
 - "no recommendation" status from FCP channel
 - invalid FSF protocol status
 - stalled outbound queue
 - unknown request id on inbound queue
 - QDIO_ERROR_SLSB_STATE

All of the above triggers run from the Linux qdio softirq context, so
no additional synchronization is necessary for the handling of the
ZFCP_STATUS_ADAPTER_SIOSL_ISSUED flag.

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@suse.de>

1 files changed, 1 insertions, 0 deletions

diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h
index 6c6374ba180..e1c6b6e05a7 100644
--- a/drivers/s390/scsi/zfcp_def.h
+++ b/drivers/s390/scsi/zfcp_def.h
@@ -73,6 +73,7 @@ struct zfcp_reqlist;
 
 /* adapter status */
 #define ZFCP_STATUS_ADAPTER_QDIOUP              0x00000002
+#define ZFCP_STATUS_ADAPTER_SIOSL_ISSUED        0x00000004
 #define ZFCP_STATUS_ADAPTER_XCONFIG_OK          0x00000008
 #define ZFCP_STATUS_ADAPTER_HOST_CON_INIT       0x00000010
 #define ZFCP_STATUS_ADAPTER_ERP_PENDING         0x00000100