[S390] cio: fix memleak in subchannel validation

When scanning for new subchannels we have a code path where we allocate
memory for a struct subchannel, set the device name (which is dynamically
allocated now) and do a check if the underlying device is blacklisted - if
so we free the subchannel structure.
Since we have not set up refcounting at this stage, the device name's memory
is lost. Fix this by moving the dev_set_name after the blacklist test.

Note: With this patch the init_name for the console subchannel becomes
virtually obsolete.

Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

drivers/s390/cio/cio.c

@@ -564,11 +564,6 @@ int cio_validate_subchannel(struct subchannel *sch, struct subchannel_id schid)
 			goto out;
 	}
 	mutex_init(&sch->reg_mutex);
-	/* Set a name for the subchannel */
-	if (cio_is_console(schid))
-		sch->dev.init_name = cio_get_console_sch_name(schid);
-	else
-		dev_set_name(&sch->dev, "0.%x.%04x", schid.ssid, schid.sch_no);
 
 	/*
 	 * The first subchannel that is not-operational (ccode==3)

drivers/s390/cio/css.c

@@ -183,6 +183,11 @@ static int css_sch_device_register(struct subchannel *sch)
 	int ret;
 
 	mutex_lock(&sch->reg_mutex);
+	if (cio_is_console(sch->schid))
+		sch->dev.init_name = cio_get_console_sch_name(sch->schid);
+	else
+		dev_set_name(&sch->dev, "0.%x.%04x", sch->schid.ssid,
+			     sch->schid.sch_no);
 	ret = device_register(&sch->dev);
 	mutex_unlock(&sch->reg_mutex);
 	return ret;