[PATCH] mpt3sas: Fix a double unlock in _transport_smp_handler()

[Dan Carpenter]

We can't "goto out;" if we're not holding "ioc->transport_cmds.mutex".
It leads to a double unlock bug, and I don't think we should set
"ioc->transport_cmds.status" if we don't have the lock.

Fixes: 651a01364994 ("scsi: scsi_transport_sas: switch to bsg-lib for SMP 
passthrough")
Signed-off-by: Dan Carpenter <dan.carpen...@oracle.com>
---
I'm not totally sure about the .status thing.  This is a static checker
fix.

diff --git a/drivers/scsi/mpt3sas/mpt3sas_transport.c 
b/drivers/scsi/mpt3sas/mpt3sas_transport.c
index d3940c5d079d..c9cd9ed90002 100644
--- a/drivers/scsi/mpt3sas/mpt3sas_transport.c
+++ b/drivers/scsi/mpt3sas/mpt3sas_transport.c
@@ -1936,12 +1936,12 @@ _transport_smp_handler(struct bsg_job *job, struct 
Scsi_Host *shost,
                pr_info(MPT3SAS_FMT "%s: host reset in progress!\n",
                    __func__, ioc->name);
                rc = -EFAULT;
-               goto out;
+               goto job_done;
        }
 
        rc = mutex_lock_interruptible(&ioc->transport_cmds.mutex);
        if (rc)
-               goto out;
+               goto job_done;
 
        if (ioc->transport_cmds.status != MPT3_CMD_NOT_USED) {
                pr_err(MPT3SAS_FMT "%s: transport_cmds in use\n", ioc->name,
@@ -2066,6 +2066,7 @@ _transport_smp_handler(struct bsg_job *job, struct 
Scsi_Host *shost,
  out:
        ioc->transport_cmds.status = MPT3_CMD_NOT_USED;
        mutex_unlock(&ioc->transport_cmds.mutex);
+ job_done:
        bsg_job_done(job, rc, reslen);
 }