USB: serial: safe-serial: fix up printk() usage

The driver was calling printk() directly at startup, which is just
noise.  Switch over to using pr_info() where needed, and get rid of the
totally useless version number that had never ever been incremented.

Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/usb/serial/safe_serial.c

@@ -62,6 +62,7 @@
  *
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/kernel.h>
 #include <linux/errno.h>
@@ -84,7 +85,6 @@
 static bool safe = 1;
 static bool padded = CONFIG_USB_SERIAL_SAFE_PADDED;
 
-#define DRIVER_VERSION "v0.1"
 #define DRIVER_AUTHOR "sl@lineo.com, tbr@lineo.com, Johan Hovold <jhovold@gmail.com>"
 #define DRIVER_DESC "USB Safe Encapsulated Serial"
 
@@ -311,13 +311,9 @@ static int __init safe_init(void)
 {
 	int i;
 
-	printk(KERN_INFO KBUILD_MODNAME ": " DRIVER_VERSION ":"
-	       DRIVER_DESC "\n");
-
 	/* if we have vendor / product parameters patch them into id list */
 	if (vendor || product) {
-		printk(KERN_INFO KBUILD_MODNAME ": vendor: %x product: %x\n",
-		       vendor, product);
+		pr_info("vendor: %x product: %x\n", vendor, product);
 
 		for (i = 0; i < ARRAY_SIZE(id_table); i++) {
 			if (!id_table[i].idVendor && !id_table[i].idProduct) {