Fnic: Fnic Driver crashed with NULL pointer reference

When issuing I/O request, if the I/O completes before returning from
fnic_queuecommand(), we may be referencing scsi_cmnd structure that may
be freed by interrupt handler. Acquring IO lock would synchronize
fnic_queuecommand and interrupt handler.

- Increment fnic version from 1.6.0.15 to 1.6.0.16

Signed-off-by: Hiral Shah <[email protected]>
Signed-off-by: Sesidhar Baddela <[email protected]>
Signed-off-by: Anil Chintalapati <[email protected]>
Signed-off-by: Christoph Hellwig <[email protected]>

drivers/scsi/fnic/fnic.h

@@ -39,7 +39,7 @@
 
 #define DRV_NAME		"fnic"
 #define DRV_DESCRIPTION		"Cisco FCoE HBA Driver"
-#define DRV_VERSION		"1.6.0.15"
+#define DRV_VERSION		"1.6.0.16"
 #define PFX			DRV_NAME ": "
 #define DFX                     DRV_NAME "%d: "
 

drivers/scsi/fnic/fnic_scsi.c

@@ -421,9 +421,10 @@ static int fnic_queuecommand_lck(struct scsi_cmnd *sc, void (*done)(struct scsi_
 	int ret;
 	u64 cmd_trace;
 	int sg_count = 0;
-	unsigned long flags;
+	unsigned long flags = 0;
 	unsigned long ptr;
 	struct fc_rport_priv *rdata;
+	spinlock_t *io_lock = NULL;
 
 	if (unlikely(fnic_chk_state_flags_locked(fnic, FNIC_FLAGS_IO_BLOCKED)))
 		return SCSI_MLQUEUE_HOST_BUSY;
@@ -509,6 +510,13 @@ static int fnic_queuecommand_lck(struct scsi_cmnd *sc, void (*done)(struct scsi_
 		}
 	}
 
+	/*
+	* Will acquire lock defore setting to IO initialized.
+	*/
+
+	io_lock = fnic_io_lock_hash(fnic, sc);
+	spin_lock_irqsave(io_lock, flags);
+
 	/* initialize rest of io_req */
 	io_req->port_id = rport->port_id;
 	io_req->start_time = jiffies;
@@ -525,11 +533,9 @@ static int fnic_queuecommand_lck(struct scsi_cmnd *sc, void (*done)(struct scsi_
 		 * In case another thread cancelled the request,
 		 * refetch the pointer under the lock.
 		 */
-		spinlock_t *io_lock = fnic_io_lock_hash(fnic, sc);
 		FNIC_TRACE(fnic_queuecommand, sc->device->host->host_no,
 			  sc->request->tag, sc, 0, 0, 0,
 			  (((u64)CMD_FLAGS(sc) << 32) | CMD_STATE(sc)));
-		spin_lock_irqsave(io_lock, flags);
 		io_req = (struct fnic_io_req *)CMD_SP(sc);
 		CMD_SP(sc) = NULL;
 		CMD_STATE(sc) = FNIC_IOREQ_CMD_COMPLETE;
@@ -538,6 +544,10 @@ static int fnic_queuecommand_lck(struct scsi_cmnd *sc, void (*done)(struct scsi_
 			fnic_release_ioreq_buf(fnic, io_req, sc);
 			mempool_free(io_req, fnic->io_req_pool);
 		}
+		atomic_dec(&fnic->in_flight);
+		/* acquire host lock before returning to SCSI */
+		spin_lock(lp->host->host_lock);
+		return ret;
 	} else {
 		atomic64_inc(&fnic_stats->io_stats.active_ios);
 		atomic64_inc(&fnic_stats->io_stats.num_ios);
@@ -559,6 +569,11 @@ static int fnic_queuecommand_lck(struct scsi_cmnd *sc, void (*done)(struct scsi_
 		  sc->request->tag, sc, io_req,
 		  sg_count, cmd_trace,
 		  (((u64)CMD_FLAGS(sc) >> 32) | CMD_STATE(sc)));
+
+	/* if only we issued IO, will we have the io lock */
+	if (CMD_FLAGS(sc) & FNIC_IO_INITIALIZED)
+		spin_unlock_irqrestore(io_lock, flags);
+
 	atomic_dec(&fnic->in_flight);
 	/* acquire host lock before returning to SCSI */
 	spin_lock(lp->host->host_lock);