[PATCH] 32-bit dev_t: md fallout

This fixes a NULL ptr dereference in md_probe() noted by Helge Hafting.

It even gets the things slightly better than they used to be, but
late-boot magic in md.c is certainly a mess.  Oh, well...

It's _ugly_.  md_probe() is misused there big way.  The minimal fix is
to revert the cleanup in md_probe() - replace

	int unit = *part;
with
	int unit = MINOR(dev);


However, that is crap solution.  The problem is that md_probe() is called
directly with bogus arguments - not only part is NULL (which triggers the
oops), but dev (which is supposed to be dev_t value) is actually mdidx(mddev).

Cleaner fix follows, but we really need to get the situation with gendisk
allocations into the sane shape there.  Sigh...

drivers/md/md.c

@@ -1500,6 +1500,7 @@ static int do_md_run(mddev_t * mddev)
 	mdk_rdev_t *rdev;
 	struct gendisk *disk;
 	char b[BDEVNAME_SIZE];
+	int unit;
 
 	if (list_empty(&mddev->disks)) {
 		MD_BUG();
@@ -1591,8 +1592,9 @@ static int do_md_run(mddev_t * mddev)
 		invalidate_bdev(rdev->bdev, 0);
 	}
 
-	md_probe(mdidx(mddev), NULL, NULL);
-	disk = disks[mdidx(mddev)];
+	unit = mdidx(mddev);
+	md_probe(0, &unit, NULL);
+	disk = disks[unit];
 	if (!disk)
 		return -ENOMEM;