init/mount: print pretty name of root device when panics

Given a wrong root device, current log may not give the pretty name
which is useful to locate root cause.

For example, there are 2 blk devs in a VM, /dev/vda which has 2 partitials
/dev/vda1 and /dev/vda2 and /dev/vdb which is blank. /dev/vda2 is the
right root dev. When set "root=/dev/vdb", we get error log:

[    0.635575] Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(254,16)

It's not straightforward to find out the root cause as there is lack of
the root devive name therefore hard for people to get those info from the
device number, in the example, (254,16).

It is more comprehensive way to hint the root cause if pretty name is
given here, like:

[    0.559887] Kernel panic - not syncing: VFS: Unable to mount root fs on "/dev/vdb" or unknown-block(254,16)

Signed-off-by: Jianyong Wu <[email protected]>
Message-Id: <[email protected]>
Signed-off-by: Christian Brauner <[email protected]>

init/do_mounts.c

@@ -244,7 +244,7 @@ void __init mount_root_generic(char *name, char *pretty_name, int flags)
 	for (i = 0, p = fs_names; i < num_fs; i++, p += strlen(p)+1)
 		printk(" %s", p);
 	printk("\n");
-	panic("VFS: Unable to mount root fs on %s", b);
+	panic("VFS: Unable to mount root fs on \"%s\" or %s", pretty_name, b);
 out:
 	put_page(page);
 }